July 31, 2023 • 44 mins
How good is your memory, really? Why do we feel so certain about our memory, even while it is so suspect? Can something told to you after an event change your memory of the event? Why is it so hard to reconstruct a face? Is there a relationship between confidence and accuracy? How is your ability to remember a scene changed if there's a gun pointed at you? Join Eagleman to find out why eyewitness testimony is the most questionable technology we allow in courts.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:04):
When you witness an event, when you see it with
your own eyes, it certainly feels like what you saw
can't be questioned. But how good is your memory? Really?
Can you misremember details? Is your memory like a video
or isn't it? Is your ability to remember changed by
(00:25):
how many things are going on in the scene, or
whether there's a gun pointed at you. Can something told
to you after the event change your memory of the event.
Welcome to Inner Cosmos with me David Eagleman. I'm a
neuroscientist and an author at Stanford and in these episodes
(00:47):
we sail deeply into our three pound universe to understand
the intersection between brains and our lives. Today's episode, we're
going to be talking about eyewitness testimony. I think it's
(01:10):
fair to say that most of us when we see
something happen, we know what we've just seen. This occurred,
this is what the person looked like, and so on.
And in fact, when jurors get interviewed on this point,
they'll often say something like, yes, I understand there are
problems with memory and eyewitness testimony, but my memory is
(01:32):
like a video recorder. But the question we're going to
ask today is how good is our memory really and
what does that mean for courts of law? So strap
in because we're going to see some amazing events and
court cases that may change your opinion on your own
memory and about what you take to be true. One
(01:58):
of the courses I teach it Anford is the Brain and
the Law. It's where neuroscience intersects with the legal system.
And a few months ago I was in the middle
of lecturing to my seventies students when something very wild happened.
I was talking to the auditorium and the back door opens,
(02:20):
and a middle aged woman comes in at the back
of the classroom, and she walks halfway down the aisle
toward me and then just interrupts me, as though I
weren't talking at all. She says, are you David Eagleman?
And I said yes, but I'm in the middle of
teaching a class. Who are you? And she starts shouting
(02:40):
how she's emailed me over and over and hasn't gotten
any reply. So I said, I'm sorry, I have an
overwhelmed inbox and I can't get to all my emails,
but I am teaching a class and I'll have to
talk with you later. But she's clearly very agitated, and
she doesn't leave, and she keeps shouting at me and
(03:00):
takes a step closer to me down the aisle and
asks if I'm ever planning to answer her, or whether
I'm going to continue to ignore her. And I say, ma'am,
you're going to have to leave and I'll talk with
you afterwards. And finally, after what seemed like forever, although
it was presumably less than a minute, she says, I'm
(03:22):
going to wait for you out here, and she goes
stomping out the back door. So I tried to maintain normalcy,
and I continued the lecture for about thirty more minutes.
And this class is a long one. I give three
hours of lecture, so normally at this point I give
my students a break at the midpoint, where we go
(03:44):
outside and we stretch for a few minutes. But I said
to the class, look, it's our break time. But I'm
not sure that it's going to make sense to go
out there. I'm going to call campus security just to
let them know about that woman, and I want to
give them a description, but I couldn't really tell her
height and wait very well from up here at the podium,
(04:07):
and I guess I was so caught up in the
surprise of the whole thing and thinking about the ways
this scenario could evolve badly that I didn't really encode
all the details that I wanted to, like what she
was wearing, or her hair color, or the details of
her face as well as I wish that I had
encoded these. So I told the class, look, I remember
(04:29):
the random fact that she had sunglasses perched on top
of her head, but I would really like your help
in getting a rich description. So I got everyone to
take out a piece of notebook paper, and I told
them the best way to accurately identify someone is to
have a collection of witnesses do this independently. So I said, look,
(04:50):
if you could draw a picture of what she looked
like and also write down her height and her weight,
then I'm going to compile these and give Stanford Security
an excellent description. So one thing I said is don't
look at each other's drawing. Just drawn on your own
and then hand it up to me. So everyone drew
their best version of how they remembered her, and some
(05:12):
people relied more on the drawing part and some on
the verbal descriptions. I collected these all up, and I said, great,
I'm going to have a very complete average version of
her that I can communicate to security. So I started
flipping through these and reading them out loud to the class,
and we were all shocked because one person described her
(05:34):
as five foot four and wearing a blue shirt and
the next described her as five foot eight and wearing
a white shirt. And the age estimates were between forty
and sixty, and the hair color was everything from brown
hair to gray hair to light hair and so on.
And the body weight ranged over about forty pounds. So
(05:59):
the whole collection provided me with no benefit at all
in calling security. So I didn't call security, but not
for the reason you might think. The reason I didn't
call is because the woman was a professional actor that
I had hired. And the whole point of the exercise
was to demonstrate firsthand to my students that vision is
(06:23):
not like a camera and memory is not like a recorder.
And from there I segued into the real topic of
the lecture, which was on the problems and challenges of
eyewitness testimony. So let's begin with a woman named Jennifer Thompson.
She was twenty two and living in North Carolina, and
(06:44):
a man broke into her apartment and raped her. It
was a long ordeal, and Jennifer really paid attention to
his face so that she would be able to identify
him later. When this whole terrible event was over and
he ran off, she drove herself to the hospital and
had a rape kit taken, and then she went straight
(07:05):
to the police, and with their help she was able
to create a composite image of what he looked like.
So the police went out and they gathered suspects, and
they put together a lineup of seven men, and Jennifer
concluded that one man in that lineup, a man named
Ronald Cotton, was the man who had raped her. So
Cotton went to jail for the crime, but the whole
(07:26):
time he was there he maintained his innocence, and many
years into his jail sense he saw the oj Simpson
trial on the prison television and he heard about DNA evidence,
which was something he had never heard of before, and
so he contacted his lawyer and he said, what is
this DNA evidence? Is this something we should look into?
(07:48):
And his lawyer was able to get a hold of
the rape kit from almost eleven years earlier, and they
were able to extract enough DNA from that, and it
turned out that Cotton was in fact innocent. He was
not the man who had done it. The man who
had raped Jennifer Thompson was a man named Bobby Poole,
and he was then apprehended and confessed to the rape.
(08:13):
And so Ronald Cotton was released after having spent eleven
years of his life in prison for a crime he
did not commit. So Jennifer was shattered by this because
she realized now that, based on her eyewitness testimony and
her certainty, she had sent an innocent man to jail
(08:33):
for eleven years. And she said that one of the
most important things to her as a rape victim was
to never feel any guilt over what had happened to her.
It wasn't her fault that she had been raped, But
all of a sudden, now she was crushed with guilt
because she had ruined an innocent man's life by erroneously
identifying him in a lineup. So, about three years after
(08:57):
Ronald Cotton got released, she decided she was going to
reach out to him, make contact with him, and of
course she was terrified about doing this. But she contacted
him anyway, and they met at a church and he
said that he had already forgiven her, and they talked
at length, and they both cried, and eventually they became friends,
(09:19):
and they ended up writing an excellent book together. It's
called Picking Cotton. And one of the things she does
in the book is wrestle with the question that was
most deeply burned into her mind. How could she have
picked the wrong person and have felt so certain about it.
How can one feel so sure and be wrong. So
(09:42):
they travel around and give talks together about the fallibility
of eyewitness testimony, about how you can believe one hundred
percent that's the face, but it doesn't actually necessitate that
you are correct. I met the two of them at
a conference when they were talking about the book. They're
close friends now, and they travel around together so much
(10:03):
that often people mistake them for a couple, and they say, oh,
how did you two meet, And they say, it's a
long story, okay. But the heart of the problem is
that memory is always a reconstruction. Memory is not like
a videotape. It's not as though your memory is something
where your visual system is taking in information and retaining
(10:25):
zeros and ones the way a cell phone video would.
All you ever see is what you believe you are
seeing out there. So if I were to ask you
the details of what's in front of you right now,
you'd be able to answer it once you pay attention
to things. But you weren't aware of that even though
it's been sitting on your retina this whole time. You're
(10:45):
not actually seeing the world like a camera. Instead, what
you have is a rough internal model of what's going on.
And when you need to take in more information, then
you do. You go out and point your eyes and
pull in more in. So when I ask you about
the scene in front of you, you need to employ
(11:05):
your attentional mechanisms to go and crawl the scene out
there and try to identify what's sitting there. So today's
question is how reliable is eyewitness testimony? In the courtroom,
for my students, I was able to demonstrate to them
the massive variety of their drawings tall, short, heavy, light,
(11:27):
curly hair, straight hair, glasses, no glasses, and none of
their drawings looked particularly like the actor at all. But
forget a classroom test, how does this play out in
the real world? How often does I witness testimony end
up convicting the wrong person, as happened with Jennifer and
Ronald Well. One thing you could do is look at
(11:48):
people who are found to be innocent years after they
were convicted for a crime, and then figure out what
percentage of them were convicted in whole or in part
based on iwold testimony. So you may have heard of
the Innocence Project. It is for people who are serving
time in prison and are maintaining their innocence. The lawyers
(12:10):
and forensic scientists at the Innocence Project go back and
open these cold cases and work to reassess the case
with DNA evidence. So what the Innocence Project found out
of the two hundred and forty people that they have exonerated,
in other words, who have been found innocent of the
crime they were convicted for, is that over sixty percent
(12:32):
of them went to jail in whole or in part
because of eyewitness testimony where someone said, I know that's
the guy. I saw him with my own eyes. There's
no doubt in my mind. Now this leads to a
question does eyewitness testimony matter? How swaying is it to jurors? Well,
the answer is it's enormously swaying. The US Supreme Court
(12:57):
Justice William Brennan pointed out that vote all the evidence
points rather strikingly to the conclusion that there is almost
nothing more convincing than a live human being who takes
the stand, points a finger at the defendants and says,
that's the one unquote. So imagine you have a prosecutor
(13:34):
showing circumstantial evidence, or a scientist comes up and says, look,
here's my neuroimaging evidence, or you have a lie detector
expert or whatever, and then you get an eyewitness and
she takes the stand and she has tears in her eyes.
She's genuine, she says, I know what I saw. He
is the one that has enormous way on jurors. Justice
(13:58):
Brennan goes on to point out that eyewent his testimony
is the type of evidence that quote, juries seem most
receptive to and not inclined to discredit end quote. In
other words, a scientist might say something on the stand
and the jury thinks, yeah, maybe, but there's another interpretation.
But when it comes to an eyewitness, we tend to say, Okay,
(14:20):
that's the gospel. Finally, Brennan points out that the court
has long recognized, at least since the nineteen sixties, the
quote inherently suspect qualities of eyewitness identification evidence unquote, and
that the court finds that kind of evidence quote notoriously
unreliable end quote. So that's Brennan, having spent a life
(14:43):
on the bench, describing what he knows to be true.
So why is it so swaying? Well, maybe we can
find some evidence from the laboratory. So my colleague, Elizabeth
loftis that you see, Irvine is probably the most famous
memory eyewitness researcher. Here's one example of her many many studies.
You tell mock jurors the details of a case. They
(15:06):
get to hear the attorneys on both sides, the prosecution
and the defense, give their arguments. Some of the jurors
only hear circumstantial evidence. There's no eyewitness testimony in that case.
And given the facts that they've heard, eighteen percent of
the mock jurors conclude that the guy is guilty. But
the other group, here's eyewitness testimony, everything else being the same,
(15:30):
and now seventy two percent of the mock jurors find
the guy guilty. So it goes from eighteen percent of
people finding him guilty to seventy two percent just based
on the eyewitness testimony. That means that even though we
know eyewitness testimony is inherently unreliable, just look at the
effect of saying, we have someone who saw that person.
(15:53):
Maybe it was a little bit far away, maybe it
was sort of dark out, but I'm pretty sure that
was the guy, and bam, it goes up to seventy two.
It's a huge difference, And for something that's so unreliable,
it's a little hard to justify that kind of pull.
So we know that eyewitness testimony is swaying, But why
is it so bad? Why can't we look at somebody
(16:16):
like the actor who broke into my classroom and just
describe accurately what she looked like. Well, for starters, it's
really really hard to remember what somebody looks like. Think
about the last time you went to the coffee shop,
what precisely did the person at the cash register look like?
Or if you had to describe your neighbor down the
(16:39):
street who you don't see that often to the police,
how well could you do so? If you really had
to pull out a piece of paper and draw the
face or describe it verbally. So police have worked for
decades to try to make this easier. The original approach
was to have a trained artist who sketches while you
(17:00):
and then you can work back and forth together. But
eventually the police put together systems where you could just
look at the individual features like the eyes and the
eyebrows and the nose and the mouth, and you flip
them until you get the face that you want out
of that. So instead of me asking you to take
out a piece of notebook paper and draw seth Rogen,
(17:21):
instead I say, look, here's a bunch of eyes and
noses and mouths, and I want you to flip these
around until you get a good picture of seth Rogen
or the person that you saw or the perpetrator. So
the first system for this was introduced in the late
nineteen fifties by the Los Angeles Police Department. It was
these features on transparent sheets and you rotate them around
(17:44):
and you reconstruct the face. This was called Identicit, and
Identicate was used by Scotland Yard in a case in
nineteen sixty one, a woman was stabbed to death in
an antique shop where she work, and the police interviewed
people and learned that someone suspicious had been in the
shop a few days earlier, and so Scotland Yard turned
(18:07):
to this invention from America and had their witnesses compile
the suspect's face using identicit, And some days later a
policeman saw a man named Edwin Bush on the street
and recognized him from the identicate pictures in the newspaper,
and in fact, when they arrested him and took him
to the station, they found he had a copy of
(18:30):
the newspaper drawing in his own pocket, and his shoes
matched the size of the shoes in the crime scene,
and so on, and he ended up confessing to the murder.
So this was considered a real success in how you
could crank up the reliability of identifying people and identic
It was eventually replaced by photo fit in the nineteen seventies,
(18:51):
where instead of using drawings, you use collages of photos
of eyes and ears and noses and mouth and hairline
and so on, and the photos give a better image
of the suspect's face rather than the line drawings. And
with due course all this evolved into software in which
you take a three dimensional model of the head and
(19:12):
you do the same sort of thing where you can
change the eyes and nose and mouth and eyebrows and
so on. And there were many stories that came out
of all these systems and that increased confidence in the approach.
But it turns out that despite success stories with this,
when you really start looking into it, it turns out
these face recognition approaches are generally not so good in
(19:33):
terms of being able to identify people. Some years ago,
an MIT researcher named Pawan Sinha got a hold of
the standard facial software that police forces use, and he
requested an expert with several years of experience with the
system to put together reconstructions directly from a picture and
(19:54):
without time constraints. That meant the operator didn't have to
rely on verbal descriptions. He could just look at the
photos and put this together. So each year in my class,
I show the students these four photographs and I offer
ten dollars to whoever can identify the pictures, and no
one ever can. There are lots of wrong guesses, but
(20:15):
so far no one has gotten them. Right. Now, the
pictures are of Bill Cosby, Carl Sagan, Ronald Reagan, Michael Jordan.
Now I get it. These are faces that the younger
generation can't necessarily identify anyway. But at the peak of
these guys popularity in the nineteen eighties, people couldn't identify
(20:35):
these pictures. Then the reconstructions just weren't close enough to
the actual face to make the match. And in case
you think this might just be a laboratory artifact, there
are plenty of real life cases where you look at
the identicate composite and then you look at the captured
person's photograph and you'll know right away that you would
(20:58):
have never identified that person from that drawing. Why. Well,
part of the problem is that the feature based composite
image is built from the pieces and parts. You say,
I think the guy's eyes looked sort of like this,
and his mouth looked sort of like that, and his
nose and his eyebrows and so on. But that's not
the way the human visual system works. It recognizes faces
(21:21):
based on the whole picture. What's called the gestalt, which
is the German word that signifies when a whole is
perceived as more than the sum of its parts. But
that is only one problem. Let's imagine that you could
somehow reconstruct a pretty good likeness there's also the problem
(21:41):
of similarity, which is to say, a lot of people
just look alike. So there was an amazing case some
years ago in which a man named Lawrence Berson, who
had curly hair and big square glasses, was convicted of
several rape cases and went to jail. And then another guy,
George Morales, with curly hair and big square glasses, was
(22:03):
convicted of several robberies and went to jail, and both
protested their innocence, and it turns out they were telling
the truth. The guy who did all of the crimes
was a guy named Richard Carbone, who had curly hair
and big square glasses. Carbone had been described by several eyewitnesses,
(22:23):
and these two other poor guys served jail time because
they were picked out of the police lineups. And putting
a photo of these guys on eagleman dot com slash
podcast because it's so striking how different people can look
similar to one another. So I've told you so far
that the brain in codes faces holistically, not in bits
and parts, and also that a lot of people look
(22:45):
roughly similar. But neither of those problems even competes with
the biggest problem of eyewitness testimony, and that is our
memories are terrible. We all like to think that we
can remember a face in describe it, but as my
students saw from trying to draw the woman who broke
into my classroom and spoke angrily with me for a minute,
(23:08):
it's really hard to remember the details of someone's face
and then try to reconstruct that. And the past few
decades have seen really great studies which undermine our confidence
that this should be easy to do. So the first
thing to appreciate is that there are two phases of memory.
The first is encoding, So when the woman burst into
(23:30):
my classroom, the students were encoding what they were seeing.
Their brains were writing it down, and the second phase
is retrieval, or pulling the memory back up later. In
the case of my classroom, I waited about thirty minutes
before asking them to draw the face, and it turns
out the forgetting curve is quite steep. By thirty minutes in,
(23:51):
they've forgotten most of the important details. So let's break
these down one at a time. We'll start with encoding,
writing down the memory. One of the biggest problems with
encoding a memory during a crime is what's called weapon focus.
So if a person has a knife on you, or
is wielding a gun, or has a baseball bat or whatever,
(24:13):
your brain can't help but to be focused on that,
and as a result, it's harder to remember details about
the person's face. So here's another study from Elizabeth loftis
she has you come into a room and she tells
you you're going to participate in some psychology exam. And
while you're waiting there, you hear two people arguing, and
(24:34):
then one person comes out of the room and he's
got some grease on his hands and he's holding a pen.
So that scenario one. Scenario two is to hear the
same thing, the two people arguing, and then the guy
comes out and he's holding a blood stained knife. And
then she studies how well you can identify the person
after that, And you can imagine what happens when the
(24:56):
guy is holding the blood stained knife. How good is
your identification about what his face looked like. It's terrible.
When there's a weapon involved, witnesses are not encoding the
details about the person. They have high anxiety and they're
staring at the weapon. It's the most salient thing in
the scene. It is the ball to keep your eye on,
(25:17):
so weapon focus is one problem. A second problem with
encoding memory during a crime scene is what is called
Q overload. When there's a really salient scene going on,
when there's lots of stuff happening, it's just hard to
encode all of that. So let's say that the woman
had come into my classroom and she was not only
yelling at me, but she starts throwing something at me.
(25:39):
And then as soon as that happens, I pull out
a taser and then a student over on the left
starts screaming, and suddenly a student on the right leaps
up and says, I'm going to protect doctor Eagleman and
dives in to tackle her. Imagine, all this stuff is happening,
bang bang bang, Well, it's really hard to encode all
that because there's just too much going on at the
same time, too many salient events happening at once. That's
(26:03):
the Q overload effect. A third problem with encoding memory
has to do with what's called the other race effect.
We're much better at encoding faces of people who look
like us, but for people who look different, it's harder
to catch the important details. Now, just to be clear,
(26:24):
this isn't racism. It's not that you're discriminating against a group.
The issue here is that the neurons in your visual
cortex just aren't trained up on the details of those faces.
It's a matter of what type of faces your brain
knows well. In other words, depending on where you've grown up,
you might have a difficult time making an accurate identification
(26:47):
if you were shown a lineup of Cambodians or Icelanders,
or mauor or Wigers or Inuits. It's not that you
have anything against these cultures. It's that you've grown training
your visual system on the faces particular to your culture,
whatever that is, and now you're dealing with measurements that
(27:07):
are all a little bit different, like the intraocular distance
and the nose length and the frenulum and the chin
shape and so on. And if that's not where your
experience and expertise lies, you're going to be worse at it.
So the other race effect often leaves eyewitness identification hobbled.
And the fourth problem with encoding memories is that it
(27:29):
often just performs worse when there's stress and trauma. During
an event, something really awful and unexpected happens and it's
simply not even part of what your world model thinks
is a possibility, and so your brain has a difficult
time encoding what the heck just happened. So these are
all problems with encoding the memory in the first place,
(27:51):
weapon focus, Q overload, other race effect, stress and trauma
during the event. Now, encoding is only half the game,
(28:14):
because retrieval is the other half. So let's say you've
written down some sort of rough memory of the woman
who came into the classroom. She's somewhere between five three
and five nine. Okay, you encoded that sort of, but
now there are problems with the retrieval of the memory,
pulling up what was there in your brain. One problem
(28:34):
is what's called the misinformation effect. If you're told something
about a crime in terms of what happened at the scene,
or who was there or what they looked like, that
will become part of your memory and you may not
be able to distinguish that from what actually happened. So
Elizabeth Loftus and others have studied this by showing people
(28:56):
a picture of a car to stop sign, and then afterwards,
after after the picture is gone, they give a text
description of the same picture, but in the text they
say it was a yield sign. And then they have
people draw the picture as they remember it, and they
draw the original scene but with a yield sign. So
they're told something after the memory was encoded. They're told
(29:19):
that it was a yield sign, and when they're retrieving
the memory, they believe that the whole time they saw
the yield sign there instead of the stop sign. In
the case of my classroom invader, you may remember that
I told the students that I remembered the woman had
sunglasses on top of her head. Well, she didn't have
sunglasses on her head. I made up that fact and
(29:41):
I asserted it, and a number of students incorporated that
false item into their memory what they think they saw,
and afterwards they felt certain that that was a part
of their original memory. Now that might sound crazy, because
you think I could distinguish my own memory from something
someone else said, but it turns out you often can't.
(30:03):
So misinformation after the fact is one problem that happens
with retrieval. A second problem is what's known as unconscious transference.
Now to explain this, I'm going to tell you an
absolutely incredible true story. There's a British psychologist named Donald Thompson,
and one evening he went on British television to talk
(30:26):
about eyewitness testimony and memory. Unbeknownst to him, while he
was on live television, a woman in England had her
apartment broken into and she was raped. She went to
the police and described Donald Thompson and had his face drawn.
She insisted this was her rapist. He was arrested the
(30:47):
next day, but he said, I have a watertight alibi,
which is that I was in a live television studio
sitting with the police commissioner. And it took a little
bit of time for all this to get worked out
and unraveled. But of course his alley was verifiable. What
happened was he was on the television set she was
getting raped, and she transferred her memory of the rapist's
(31:10):
face to his face. She got confused about whose face
was who. It's such a crazy irony that he is
a memory expert, because it could have been anybody on
the television, but there it is incredible but true. So
unconscious transference is a problem where a victim can't distinguish
(31:31):
between the perpetrator of a crime and some other face
that they saw in the same context or a totally
different context. So those are some of the problems with
memory retrieval. And one of the places where these problems
come up all the time is in the police lineup.
So imagine that you are presented a lineup with several
(31:52):
people and you have to make a choice about which
person you saw doing the crime. Well, one of the
things that started getting recognized in the nineteen sixties and
went to the Supreme Court was the idea of police suggestibility.
It turns out that if the police already have their
man in mind, they already think it's Fred. Whether or
(32:14):
not that's correct, they believe it's Fred, and they want
you to say it's Fred. There are all kinds of
ways that they can suggest that to you, including just
things like positive feedback. So if you say I think
that was the guy, they'll say, yeah, good job, that's
what we think also, And it turns out that influences
(32:35):
the confidence of the eyewitness. When the trial starts months later,
you'll say I'm absolutely certain that it was Fred, even
though you might not remember that you weren't certain at all,
But because of the positive feedback, which can even be
quite subtle, like you know, just a nod or a
smile or whatever, your confidence goes way up. And thirty
(32:57):
years of psychology studies in the laboratory have verified the
power of this suggestibility. As a result, psychologists have made
suggestions to police forces and this has spun all the
way up to the Supreme Court, with the result that
police are not allowed to have any suggestibility involved. One
(33:17):
of the ways to take care of this is to
make the lineup identification double blind. That means the police
officer who is running the lineup doesn't even know who
the main suspect is. And this way the person doing
the identifying gets no feedback at all. And by the way,
the main thing that everyone's worried about with lineups is
false identification. If the perpetrator is in the lineup and
(33:39):
you miss him, that's another kind of problem. But the
really terrible problem is sending an innocent person to prison
with the false belief that you have identified him. I'll
give you another problem that psychologists and legal theorists have studied,
and that's the issue of co witness contamination. The idea
is that if you see a crime and I'm standing
(34:00):
there and I see it too, and then we start
talking about it with one another, we can't help but
influence each other's memories. This is a close relative of
the misinformation problem. If you remember that she had curly hair,
but I say I'm pretty sure she had straight hair.
Or if you think she was unathletic and I say, no,
she was quite athletic, both our memories become contaminated by
(34:22):
the other one statement, and we more and more come
to believe things that we didn't originally. So one thing
that police know to do straight away is to separate witnesses.
And there are many many aspects that have come from
research that are now built into the way that police
optimally do lineups. For example, they try to make sure
that you don't see photographs of a suspect before the lineup,
(34:45):
because if you do, you're really likely to identify that guy.
One place this comes up is contamination by photos in
news stories. So CNN says the suspect looks like this,
and then you're brought into a lineup, and whether or
not you remember having seen that story on CNN, that
influences your performance in the lineup. So many guidelines have
(35:07):
come out of this research. I'll give you just a
few examples. One guideline is to make sure that the
eyewitness is aware that the perpetrator might not be in
the lineup. Another is this double blind procedure that I
mentioned that doesn't allow police to see the lineup, so
they can't subject to the eyewitness to any of their
suspicions as to who the suspect is, and courts are
(35:29):
increasingly recognizing that this speed of recognition matters. Generally speaking,
if the witness quickly identifies the perpetrator, then the selection
is more likely to be correct. Finally, if the appearance
of a person stands out amongst the otherwise undistinctive crowd,
then an eyewitness is more likely to select that person,
(35:51):
regardless of their own recollection of the criminal. This is
known as the distractor or dud effect. So there are
many places where scientific study has made d contributions to
the optimal way to run. I witnessed identification procedures. Now
what if somebody comes in and they say, I saw
the guy, and I know with one hundred percent certainty
(36:12):
that was the guy, Versus someone else who comes in
and says, I don't know, I was looking out a window.
It was a car robbery going on two stories below me.
It was kind of dark. I couldn't really see him
very well. I think that was the guy. I don't
really know. Do you think those two scenarios should be
treated differently? In other words, what is the relationship between
confidence and accuracy? Well? The United States Supreme Court had
(36:36):
to answer this question in nineteen seventy two in a
case called Neil versus Biggers. What happened was that the victim,
Margaret Beemer, was dragged into the woods and assaulted. She
described her attacker to the police only in very general terms.
Then she was shown a bunch of photographs and people
who met the description, but she couldn't identify anyone from
(36:57):
the photos or stand ups for seven months. So the
police then get a guy on another charge, a got
named Archie Biggers, and they decide to include him in
a stand up, But they couldn't find anyone who made
a good match to this guy's looks, so they did
what's called a show up instead, where there's no one
else there but the one suspect. So two detectives walk
(37:20):
Biggers passed the victim, and at her request, the police
directed him to say shut up or I will kill you.
The testimony a trial wasn't clear as to whether she
had first identified him and then asked that he repeat
those words, or made her identification after he'd spoken. In
any event, the victim testified that she had no doubt
(37:42):
about her identification, and so Biggers was convicted, and he
came back on appeal and argued it simply wasn't fair,
this was unreliable. The lower courts agreed with him and said,
no way, We're not accepting this as evidence because this
identification process was extremely suggestive. But the US Supreme Court
(38:03):
eventually heard this case and reversed that decision. They concluded
it was okay why because she said that she was certain.
In other words, they concluded that how confident a person
is relates to how good their evidence is. They felt
that high confidence allows you to judge the veracity of
(38:24):
a witness. So, in other words, the Supreme Court said
implicitly that they think there is a relationship between confidence
and accuracy. But generally, cognitive psychologists find that this relationship
is very weak. Because of all the things we've talked
about in this episode. Confidence does not work as a
(38:45):
yardstick of accuracy, especially as time goes on and I'm
going to do an episode soon about how the legal
system decides which technologies to allow into their courtrooms, But
for now, I'm just going to point out that it
requires passing high scientific standards. So how do you think
eyewitness testimony compares well? A version of this question hit
(39:09):
the US Supreme Court in twenty eleven Perry versus New Hampshire.
I spoke with Sanjay Gupta on CNN about this case
when it was getting decided. The question the court was
asking was essentially, if there's lousy eyewitness testimony in a case,
should it be allowed? When does that violate the right
to do process? You'll write to a fair trial. So
(39:33):
Perry was a young man in New Hampshire who got
convicted of stealing a car, and he was convicted based
in large part on a woman who was looking out
a second story window in the dark. She couldn't see
well from that distance. So Perry got convicted in a
New Hampshire court, and they came back and tried to appeal,
and he said, how can you possibly take this woman's
(39:54):
word for it? How could you send me to jail
based on such unreliable witness testimony? The lower courts upheld
the decision, and so this question spun up to the
US Supreme Court. The question that was on the table
is does unreliable eyewitness testimony violate your constitutional rights? And
(40:15):
what the court said is, look, the only thing we
care about is whether there was suggestibility by the police.
If the police manipulated the process, then the due process
clause of the Constitution is applicable. But they said, this
does not hold generally for eyewinness testimony. Using unreliable evidence
(40:35):
against you is not unconstitutional. There's no right that you'll
have great evidence. Instead, it's the job of the jury
to figure it out, to weigh the evidence. So that's
what the court decided. And reading between the lines, what
do you think one of the concerns of the court
was here, It's this, if you introduce a question about
(40:57):
the reliability of the witness, then you would, in theory,
require a pre trial hearing to determine if her testimony
is acceptable or not. And what does that do to
every single case running in the nation. One of these
Supreme Court justices posed this during the case. He said,
what about jailhouse testimony where someone in the jail says,
(41:19):
my cellmate told me x Y and Z. Sometimes it's unreliable,
sometimes it yields fruit. Do we require a pre trial
hearing every time because it might not be reliable? So
this is an issue that would overturn the way the
whole court system works. And so, at least for the
time being, the legal system knows that eyewitness testimony is terrible.
(41:45):
It can be very unreliable, and yet with all the caveats,
there's no choice but to let it remain. So in
closing your memory is not necessarily what you think it is.
Memory is not writing down what happened in zeros and
ones and then reading that back out. There are many
(42:05):
things that get in the way of good encoding and retrieval.
I really recommend Jennifer Thompson and Ronald Cotton's book. For Jennifer,
this was a complete pouring out of her guts on
the table a cathartic enterprise where she had to allow
that she put a man in jail for eleven years
based on her certainty that that was the face she
(42:27):
had looked into. She knew it was Ronald Cotton, and
she wanted him to die. That's how she felt about
this man. And then she found out that she was wrong.
So the book is about what it's like from the
inside to believe, you know, and all the steps that
happened along the way that manipulated her memory. So to
summarize what we talked about today, eyewitness testimony has a
(42:49):
terrific sway on jurors, and yet it is variable in
its accuracy. Now, just for clarity, it's not that eyewitness
testimony always has to be wrong. Plenty of times someone
IDs someone and it's correct. But memory is not like
a video camera. It is a reconstruction, and many factors
worse than the encoding and manipulate the retrieval. As a result,
(43:13):
we cannot say that confidence and accuracy are tightly linked,
especially as time goes on. Eyewitness testimony is not going
away from the courtroom because often it's the only evidence
that we can bring to bear in a case. So,
even though it may be the worst technology that we
allow in the courtrooms, it is presumably here to stay.
(43:35):
But remember, like much about our perception of the world
that I'll be discussing throughout these episodes, just because you
believe something to be true doesn't necessitate that it is.
Go to Eagleman dot Com, Slash podcast for more information
(43:56):
and to find further readings and to see some photographs.
Send me in email at podcasts at eagleman dot com
with questions or discussion, and I'll be making an episode
soon in which I address those. Until next time, I'm
Dave Eagleman, and this is Inner Cosmos
When you witness an event, when you see it with
your own eyes, it certainly feels like what you saw
can't be questioned. But how good is your memory? Really?
Can you misremember details? Is your memory like a video
or isn't it? Is your ability to remember changed by
(00:25):
how many things are going on in the scene, or
whether there's a gun pointed at you. Can something told
to you after the event change your memory of the event.
Welcome to Inner Cosmos with me David Eagleman. I'm a
neuroscientist and an author at Stanford and in these episodes
(00:47):
we sail deeply into our three pound universe to understand
the intersection between brains and our lives. Today's episode, we're
going to be talking about eyewitness testimony. I think it's
(01:10):
fair to say that most of us when we see
something happen, we know what we've just seen. This occurred,
this is what the person looked like, and so on.
And in fact, when jurors get interviewed on this point,
they'll often say something like, yes, I understand there are
problems with memory and eyewitness testimony, but my memory is
(01:32):
like a video recorder. But the question we're going to
ask today is how good is our memory really and
what does that mean for courts of law? So strap
in because we're going to see some amazing events and
court cases that may change your opinion on your own
memory and about what you take to be true. One
(01:58):
of the courses I teach it Anford is the Brain and
the Law. It's where neuroscience intersects with the legal system.
And a few months ago I was in the middle
of lecturing to my seventies students when something very wild happened.
I was talking to the auditorium and the back door opens,
(02:20):
and a middle aged woman comes in at the back
of the classroom, and she walks halfway down the aisle
toward me and then just interrupts me, as though I
weren't talking at all. She says, are you David Eagleman?
And I said yes, but I'm in the middle of
teaching a class. Who are you? And she starts shouting
(02:40):
how she's emailed me over and over and hasn't gotten
any reply. So I said, I'm sorry, I have an
overwhelmed inbox and I can't get to all my emails,
but I am teaching a class and I'll have to
talk with you later. But she's clearly very agitated, and
she doesn't leave, and she keeps shouting at me and
(03:00):
takes a step closer to me down the aisle and
asks if I'm ever planning to answer her, or whether
I'm going to continue to ignore her. And I say, ma'am,
you're going to have to leave and I'll talk with
you afterwards. And finally, after what seemed like forever, although
it was presumably less than a minute, she says, I'm
(03:22):
going to wait for you out here, and she goes
stomping out the back door. So I tried to maintain normalcy,
and I continued the lecture for about thirty more minutes.
And this class is a long one. I give three
hours of lecture, so normally at this point I give
my students a break at the midpoint, where we go
(03:44):
outside and we stretch for a few minutes. But I said
to the class, look, it's our break time. But I'm
not sure that it's going to make sense to go
out there. I'm going to call campus security just to
let them know about that woman, and I want to
give them a description, but I couldn't really tell her
height and wait very well from up here at the podium,
(04:07):
and I guess I was so caught up in the
surprise of the whole thing and thinking about the ways
this scenario could evolve badly that I didn't really encode
all the details that I wanted to, like what she
was wearing, or her hair color, or the details of
her face as well as I wish that I had
encoded these. So I told the class, look, I remember
(04:29):
the random fact that she had sunglasses perched on top
of her head, but I would really like your help
in getting a rich description. So I got everyone to
take out a piece of notebook paper, and I told
them the best way to accurately identify someone is to
have a collection of witnesses do this independently. So I said, look,
(04:50):
if you could draw a picture of what she looked
like and also write down her height and her weight,
then I'm going to compile these and give Stanford Security
an excellent description. So one thing I said is don't
look at each other's drawing. Just drawn on your own
and then hand it up to me. So everyone drew
their best version of how they remembered her, and some
(05:12):
people relied more on the drawing part and some on
the verbal descriptions. I collected these all up, and I said, great,
I'm going to have a very complete average version of
her that I can communicate to security. So I started
flipping through these and reading them out loud to the class,
and we were all shocked because one person described her
(05:34):
as five foot four and wearing a blue shirt and
the next described her as five foot eight and wearing
a white shirt. And the age estimates were between forty
and sixty, and the hair color was everything from brown
hair to gray hair to light hair and so on.
And the body weight ranged over about forty pounds. So
(05:59):
the whole collection provided me with no benefit at all
in calling security. So I didn't call security, but not
for the reason you might think. The reason I didn't
call is because the woman was a professional actor that
I had hired. And the whole point of the exercise
was to demonstrate firsthand to my students that vision is
(06:23):
not like a camera and memory is not like a recorder.
And from there I segued into the real topic of
the lecture, which was on the problems and challenges of
eyewitness testimony. So let's begin with a woman named Jennifer Thompson.
She was twenty two and living in North Carolina, and
(06:44):
a man broke into her apartment and raped her. It
was a long ordeal, and Jennifer really paid attention to
his face so that she would be able to identify
him later. When this whole terrible event was over and
he ran off, she drove herself to the hospital and
had a rape kit taken, and then she went straight
(07:05):
to the police, and with their help she was able
to create a composite image of what he looked like.
So the police went out and they gathered suspects, and
they put together a lineup of seven men, and Jennifer
concluded that one man in that lineup, a man named
Ronald Cotton, was the man who had raped her. So
Cotton went to jail for the crime, but the whole
(07:26):
time he was there he maintained his innocence, and many
years into his jail sense he saw the oj Simpson
trial on the prison television and he heard about DNA evidence,
which was something he had never heard of before, and
so he contacted his lawyer and he said, what is
this DNA evidence? Is this something we should look into?
(07:48):
And his lawyer was able to get a hold of
the rape kit from almost eleven years earlier, and they
were able to extract enough DNA from that, and it
turned out that Cotton was in fact innocent. He was
not the man who had done it. The man who
had raped Jennifer Thompson was a man named Bobby Poole,
and he was then apprehended and confessed to the rape.
(08:13):
And so Ronald Cotton was released after having spent eleven
years of his life in prison for a crime he
did not commit. So Jennifer was shattered by this because
she realized now that, based on her eyewitness testimony and
her certainty, she had sent an innocent man to jail
(08:33):
for eleven years. And she said that one of the
most important things to her as a rape victim was
to never feel any guilt over what had happened to her.
It wasn't her fault that she had been raped, But
all of a sudden, now she was crushed with guilt
because she had ruined an innocent man's life by erroneously
identifying him in a lineup. So, about three years after
(08:57):
Ronald Cotton got released, she decided she was going to
reach out to him, make contact with him, and of
course she was terrified about doing this. But she contacted
him anyway, and they met at a church and he
said that he had already forgiven her, and they talked
at length, and they both cried, and eventually they became friends,
(09:19):
and they ended up writing an excellent book together. It's
called Picking Cotton. And one of the things she does
in the book is wrestle with the question that was
most deeply burned into her mind. How could she have
picked the wrong person and have felt so certain about it.
How can one feel so sure and be wrong. So
(09:42):
they travel around and give talks together about the fallibility
of eyewitness testimony, about how you can believe one hundred
percent that's the face, but it doesn't actually necessitate that
you are correct. I met the two of them at
a conference when they were talking about the book. They're
close friends now, and they travel around together so much
(10:03):
that often people mistake them for a couple, and they say, oh,
how did you two meet, And they say, it's a
long story, okay. But the heart of the problem is
that memory is always a reconstruction. Memory is not like
a videotape. It's not as though your memory is something
where your visual system is taking in information and retaining
(10:25):
zeros and ones the way a cell phone video would.
All you ever see is what you believe you are
seeing out there. So if I were to ask you
the details of what's in front of you right now,
you'd be able to answer it once you pay attention
to things. But you weren't aware of that even though
it's been sitting on your retina this whole time. You're
(10:45):
not actually seeing the world like a camera. Instead, what
you have is a rough internal model of what's going on.
And when you need to take in more information, then
you do. You go out and point your eyes and
pull in more in. So when I ask you about
the scene in front of you, you need to employ
(11:05):
your attentional mechanisms to go and crawl the scene out
there and try to identify what's sitting there. So today's
question is how reliable is eyewitness testimony? In the courtroom,
for my students, I was able to demonstrate to them
the massive variety of their drawings tall, short, heavy, light,
(11:27):
curly hair, straight hair, glasses, no glasses, and none of
their drawings looked particularly like the actor at all. But
forget a classroom test, how does this play out in
the real world? How often does I witness testimony end
up convicting the wrong person, as happened with Jennifer and
Ronald Well. One thing you could do is look at
(11:48):
people who are found to be innocent years after they
were convicted for a crime, and then figure out what
percentage of them were convicted in whole or in part
based on iwold testimony. So you may have heard of
the Innocence Project. It is for people who are serving
time in prison and are maintaining their innocence. The lawyers
(12:10):
and forensic scientists at the Innocence Project go back and
open these cold cases and work to reassess the case
with DNA evidence. So what the Innocence Project found out
of the two hundred and forty people that they have exonerated,
in other words, who have been found innocent of the
crime they were convicted for, is that over sixty percent
(12:32):
of them went to jail in whole or in part
because of eyewitness testimony where someone said, I know that's
the guy. I saw him with my own eyes. There's
no doubt in my mind. Now this leads to a
question does eyewitness testimony matter? How swaying is it to jurors? Well,
the answer is it's enormously swaying. The US Supreme Court
(12:57):
Justice William Brennan pointed out that vote all the evidence
points rather strikingly to the conclusion that there is almost
nothing more convincing than a live human being who takes
the stand, points a finger at the defendants and says,
that's the one unquote. So imagine you have a prosecutor
(13:34):
showing circumstantial evidence, or a scientist comes up and says, look,
here's my neuroimaging evidence, or you have a lie detector
expert or whatever, and then you get an eyewitness and
she takes the stand and she has tears in her eyes.
She's genuine, she says, I know what I saw. He
is the one that has enormous way on jurors. Justice
(13:58):
Brennan goes on to point out that eyewent his testimony
is the type of evidence that quote, juries seem most
receptive to and not inclined to discredit end quote. In
other words, a scientist might say something on the stand
and the jury thinks, yeah, maybe, but there's another interpretation.
But when it comes to an eyewitness, we tend to say, Okay,
(14:20):
that's the gospel. Finally, Brennan points out that the court
has long recognized, at least since the nineteen sixties, the
quote inherently suspect qualities of eyewitness identification evidence unquote, and
that the court finds that kind of evidence quote notoriously
unreliable end quote. So that's Brennan, having spent a life
(14:43):
on the bench, describing what he knows to be true.
So why is it so swaying? Well, maybe we can
find some evidence from the laboratory. So my colleague, Elizabeth
loftis that you see, Irvine is probably the most famous
memory eyewitness researcher. Here's one example of her many many studies.
You tell mock jurors the details of a case. They
(15:06):
get to hear the attorneys on both sides, the prosecution
and the defense, give their arguments. Some of the jurors
only hear circumstantial evidence. There's no eyewitness testimony in that case.
And given the facts that they've heard, eighteen percent of
the mock jurors conclude that the guy is guilty. But
the other group, here's eyewitness testimony, everything else being the same,
(15:30):
and now seventy two percent of the mock jurors find
the guy guilty. So it goes from eighteen percent of
people finding him guilty to seventy two percent just based
on the eyewitness testimony. That means that even though we
know eyewitness testimony is inherently unreliable, just look at the
effect of saying, we have someone who saw that person.
(15:53):
Maybe it was a little bit far away, maybe it
was sort of dark out, but I'm pretty sure that
was the guy, and bam, it goes up to seventy two.
It's a huge difference, And for something that's so unreliable,
it's a little hard to justify that kind of pull.
So we know that eyewitness testimony is swaying, But why
is it so bad? Why can't we look at somebody
(16:16):
like the actor who broke into my classroom and just
describe accurately what she looked like. Well, for starters, it's
really really hard to remember what somebody looks like. Think
about the last time you went to the coffee shop,
what precisely did the person at the cash register look like?
Or if you had to describe your neighbor down the
(16:39):
street who you don't see that often to the police,
how well could you do so? If you really had
to pull out a piece of paper and draw the
face or describe it verbally. So police have worked for
decades to try to make this easier. The original approach
was to have a trained artist who sketches while you
(17:00):
and then you can work back and forth together. But
eventually the police put together systems where you could just
look at the individual features like the eyes and the
eyebrows and the nose and the mouth, and you flip
them until you get the face that you want out
of that. So instead of me asking you to take
out a piece of notebook paper and draw seth Rogen,
(17:21):
instead I say, look, here's a bunch of eyes and
noses and mouths, and I want you to flip these
around until you get a good picture of seth Rogen
or the person that you saw or the perpetrator. So
the first system for this was introduced in the late
nineteen fifties by the Los Angeles Police Department. It was
these features on transparent sheets and you rotate them around
(17:44):
and you reconstruct the face. This was called Identicit, and
Identicate was used by Scotland Yard in a case in
nineteen sixty one, a woman was stabbed to death in
an antique shop where she work, and the police interviewed
people and learned that someone suspicious had been in the
shop a few days earlier, and so Scotland Yard turned
(18:07):
to this invention from America and had their witnesses compile
the suspect's face using identicit, And some days later a
policeman saw a man named Edwin Bush on the street
and recognized him from the identicate pictures in the newspaper,
and in fact, when they arrested him and took him
to the station, they found he had a copy of
(18:30):
the newspaper drawing in his own pocket, and his shoes
matched the size of the shoes in the crime scene,
and so on, and he ended up confessing to the murder.
So this was considered a real success in how you
could crank up the reliability of identifying people and identic
It was eventually replaced by photo fit in the nineteen seventies,
(18:51):
where instead of using drawings, you use collages of photos
of eyes and ears and noses and mouth and hairline
and so on, and the photos give a better image
of the suspect's face rather than the line drawings. And
with due course all this evolved into software in which
you take a three dimensional model of the head and
(19:12):
you do the same sort of thing where you can
change the eyes and nose and mouth and eyebrows and
so on. And there were many stories that came out
of all these systems and that increased confidence in the approach.
But it turns out that despite success stories with this,
when you really start looking into it, it turns out
these face recognition approaches are generally not so good in
(19:33):
terms of being able to identify people. Some years ago,
an MIT researcher named Pawan Sinha got a hold of
the standard facial software that police forces use, and he
requested an expert with several years of experience with the
system to put together reconstructions directly from a picture and
(19:54):
without time constraints. That meant the operator didn't have to
rely on verbal descriptions. He could just look at the
photos and put this together. So each year in my class,
I show the students these four photographs and I offer
ten dollars to whoever can identify the pictures, and no
one ever can. There are lots of wrong guesses, but
(20:15):
so far no one has gotten them. Right. Now, the
pictures are of Bill Cosby, Carl Sagan, Ronald Reagan, Michael Jordan.
Now I get it. These are faces that the younger
generation can't necessarily identify anyway. But at the peak of
these guys popularity in the nineteen eighties, people couldn't identify
(20:35):
these pictures. Then the reconstructions just weren't close enough to
the actual face to make the match. And in case
you think this might just be a laboratory artifact, there
are plenty of real life cases where you look at
the identicate composite and then you look at the captured
person's photograph and you'll know right away that you would
(20:58):
have never identified that person from that drawing. Why. Well,
part of the problem is that the feature based composite
image is built from the pieces and parts. You say,
I think the guy's eyes looked sort of like this,
and his mouth looked sort of like that, and his
nose and his eyebrows and so on. But that's not
the way the human visual system works. It recognizes faces
(21:21):
based on the whole picture. What's called the gestalt, which
is the German word that signifies when a whole is
perceived as more than the sum of its parts. But
that is only one problem. Let's imagine that you could
somehow reconstruct a pretty good likeness there's also the problem
(21:41):
of similarity, which is to say, a lot of people
just look alike. So there was an amazing case some
years ago in which a man named Lawrence Berson, who
had curly hair and big square glasses, was convicted of
several rape cases and went to jail. And then another guy,
George Morales, with curly hair and big square glasses, was
(22:03):
convicted of several robberies and went to jail, and both
protested their innocence, and it turns out they were telling
the truth. The guy who did all of the crimes
was a guy named Richard Carbone, who had curly hair
and big square glasses. Carbone had been described by several eyewitnesses,
(22:23):
and these two other poor guys served jail time because
they were picked out of the police lineups. And putting
a photo of these guys on eagleman dot com slash
podcast because it's so striking how different people can look
similar to one another. So I've told you so far
that the brain in codes faces holistically, not in bits
and parts, and also that a lot of people look
(22:45):
roughly similar. But neither of those problems even competes with
the biggest problem of eyewitness testimony, and that is our
memories are terrible. We all like to think that we
can remember a face in describe it, but as my
students saw from trying to draw the woman who broke
into my classroom and spoke angrily with me for a minute,
(23:08):
it's really hard to remember the details of someone's face
and then try to reconstruct that. And the past few
decades have seen really great studies which undermine our confidence
that this should be easy to do. So the first
thing to appreciate is that there are two phases of memory.
The first is encoding, So when the woman burst into
(23:30):
my classroom, the students were encoding what they were seeing.
Their brains were writing it down, and the second phase
is retrieval, or pulling the memory back up later. In
the case of my classroom, I waited about thirty minutes
before asking them to draw the face, and it turns
out the forgetting curve is quite steep. By thirty minutes in,
(23:51):
they've forgotten most of the important details. So let's break
these down one at a time. We'll start with encoding,
writing down the memory. One of the biggest problems with
encoding a memory during a crime is what's called weapon focus.
So if a person has a knife on you, or
is wielding a gun, or has a baseball bat or whatever,
(24:13):
your brain can't help but to be focused on that,
and as a result, it's harder to remember details about
the person's face. So here's another study from Elizabeth loftis
she has you come into a room and she tells
you you're going to participate in some psychology exam. And
while you're waiting there, you hear two people arguing, and
(24:34):
then one person comes out of the room and he's
got some grease on his hands and he's holding a pen.
So that scenario one. Scenario two is to hear the
same thing, the two people arguing, and then the guy
comes out and he's holding a blood stained knife. And
then she studies how well you can identify the person
after that, And you can imagine what happens when the
(24:56):
guy is holding the blood stained knife. How good is
your identification about what his face looked like. It's terrible.
When there's a weapon involved, witnesses are not encoding the
details about the person. They have high anxiety and they're
staring at the weapon. It's the most salient thing in
the scene. It is the ball to keep your eye on,
(25:17):
so weapon focus is one problem. A second problem with
encoding memory during a crime scene is what is called
Q overload. When there's a really salient scene going on,
when there's lots of stuff happening, it's just hard to
encode all of that. So let's say that the woman
had come into my classroom and she was not only
yelling at me, but she starts throwing something at me.
(25:39):
And then as soon as that happens, I pull out
a taser and then a student over on the left
starts screaming, and suddenly a student on the right leaps
up and says, I'm going to protect doctor Eagleman and
dives in to tackle her. Imagine, all this stuff is happening,
bang bang bang, Well, it's really hard to encode all
that because there's just too much going on at the
same time, too many salient events happening at once. That's
(26:03):
the Q overload effect. A third problem with encoding memory
has to do with what's called the other race effect.
We're much better at encoding faces of people who look
like us, but for people who look different, it's harder
to catch the important details. Now, just to be clear,
(26:24):
this isn't racism. It's not that you're discriminating against a group.
The issue here is that the neurons in your visual
cortex just aren't trained up on the details of those faces.
It's a matter of what type of faces your brain
knows well. In other words, depending on where you've grown up,
you might have a difficult time making an accurate identification
(26:47):
if you were shown a lineup of Cambodians or Icelanders,
or mauor or Wigers or Inuits. It's not that you
have anything against these cultures. It's that you've grown training
your visual system on the faces particular to your culture,
whatever that is, and now you're dealing with measurements that
(27:07):
are all a little bit different, like the intraocular distance
and the nose length and the frenulum and the chin
shape and so on. And if that's not where your
experience and expertise lies, you're going to be worse at it.
So the other race effect often leaves eyewitness identification hobbled.
And the fourth problem with encoding memories is that it
(27:29):
often just performs worse when there's stress and trauma. During
an event, something really awful and unexpected happens and it's
simply not even part of what your world model thinks
is a possibility, and so your brain has a difficult
time encoding what the heck just happened. So these are
all problems with encoding the memory in the first place,
(27:51):
weapon focus, Q overload, other race effect, stress and trauma
during the event. Now, encoding is only half the game,
(28:14):
because retrieval is the other half. So let's say you've
written down some sort of rough memory of the woman
who came into the classroom. She's somewhere between five three
and five nine. Okay, you encoded that sort of, but
now there are problems with the retrieval of the memory,
pulling up what was there in your brain. One problem
(28:34):
is what's called the misinformation effect. If you're told something
about a crime in terms of what happened at the scene,
or who was there or what they looked like, that
will become part of your memory and you may not
be able to distinguish that from what actually happened. So
Elizabeth Loftus and others have studied this by showing people
(28:56):
a picture of a car to stop sign, and then afterwards,
after after the picture is gone, they give a text
description of the same picture, but in the text they
say it was a yield sign. And then they have
people draw the picture as they remember it, and they
draw the original scene but with a yield sign. So
they're told something after the memory was encoded. They're told
(29:19):
that it was a yield sign, and when they're retrieving
the memory, they believe that the whole time they saw
the yield sign there instead of the stop sign. In
the case of my classroom invader, you may remember that
I told the students that I remembered the woman had
sunglasses on top of her head. Well, she didn't have
sunglasses on her head. I made up that fact and
(29:41):
I asserted it, and a number of students incorporated that
false item into their memory what they think they saw,
and afterwards they felt certain that that was a part
of their original memory. Now that might sound crazy, because
you think I could distinguish my own memory from something
someone else said, but it turns out you often can't.
(30:03):
So misinformation after the fact is one problem that happens
with retrieval. A second problem is what's known as unconscious transference.
Now to explain this, I'm going to tell you an
absolutely incredible true story. There's a British psychologist named Donald Thompson,
and one evening he went on British television to talk
(30:26):
about eyewitness testimony and memory. Unbeknownst to him, while he
was on live television, a woman in England had her
apartment broken into and she was raped. She went to
the police and described Donald Thompson and had his face drawn.
She insisted this was her rapist. He was arrested the
(30:47):
next day, but he said, I have a watertight alibi,
which is that I was in a live television studio
sitting with the police commissioner. And it took a little
bit of time for all this to get worked out
and unraveled. But of course his alley was verifiable. What
happened was he was on the television set she was
getting raped, and she transferred her memory of the rapist's
(31:10):
face to his face. She got confused about whose face
was who. It's such a crazy irony that he is
a memory expert, because it could have been anybody on
the television, but there it is incredible but true. So
unconscious transference is a problem where a victim can't distinguish
(31:31):
between the perpetrator of a crime and some other face
that they saw in the same context or a totally
different context. So those are some of the problems with
memory retrieval. And one of the places where these problems
come up all the time is in the police lineup.
So imagine that you are presented a lineup with several
(31:52):
people and you have to make a choice about which
person you saw doing the crime. Well, one of the
things that started getting recognized in the nineteen sixties and
went to the Supreme Court was the idea of police suggestibility.
It turns out that if the police already have their
man in mind, they already think it's Fred. Whether or
(32:14):
not that's correct, they believe it's Fred, and they want
you to say it's Fred. There are all kinds of
ways that they can suggest that to you, including just
things like positive feedback. So if you say I think
that was the guy, they'll say, yeah, good job, that's
what we think also, And it turns out that influences
(32:35):
the confidence of the eyewitness. When the trial starts months later,
you'll say I'm absolutely certain that it was Fred, even
though you might not remember that you weren't certain at all,
But because of the positive feedback, which can even be
quite subtle, like you know, just a nod or a
smile or whatever, your confidence goes way up. And thirty
(32:57):
years of psychology studies in the laboratory have verified the
power of this suggestibility. As a result, psychologists have made
suggestions to police forces and this has spun all the
way up to the Supreme Court, with the result that
police are not allowed to have any suggestibility involved. One
(33:17):
of the ways to take care of this is to
make the lineup identification double blind. That means the police
officer who is running the lineup doesn't even know who
the main suspect is. And this way the person doing
the identifying gets no feedback at all. And by the way,
the main thing that everyone's worried about with lineups is
false identification. If the perpetrator is in the lineup and
(33:39):
you miss him, that's another kind of problem. But the
really terrible problem is sending an innocent person to prison
with the false belief that you have identified him. I'll
give you another problem that psychologists and legal theorists have studied,
and that's the issue of co witness contamination. The idea
is that if you see a crime and I'm standing
(34:00):
there and I see it too, and then we start
talking about it with one another, we can't help but
influence each other's memories. This is a close relative of
the misinformation problem. If you remember that she had curly hair,
but I say I'm pretty sure she had straight hair.
Or if you think she was unathletic and I say, no,
she was quite athletic, both our memories become contaminated by
(34:22):
the other one statement, and we more and more come
to believe things that we didn't originally. So one thing
that police know to do straight away is to separate witnesses.
And there are many many aspects that have come from
research that are now built into the way that police
optimally do lineups. For example, they try to make sure
that you don't see photographs of a suspect before the lineup,
(34:45):
because if you do, you're really likely to identify that guy.
One place this comes up is contamination by photos in
news stories. So CNN says the suspect looks like this,
and then you're brought into a lineup, and whether or
not you remember having seen that story on CNN, that
influences your performance in the lineup. So many guidelines have
(35:07):
come out of this research. I'll give you just a
few examples. One guideline is to make sure that the
eyewitness is aware that the perpetrator might not be in
the lineup. Another is this double blind procedure that I
mentioned that doesn't allow police to see the lineup, so
they can't subject to the eyewitness to any of their
suspicions as to who the suspect is, and courts are
(35:29):
increasingly recognizing that this speed of recognition matters. Generally speaking,
if the witness quickly identifies the perpetrator, then the selection
is more likely to be correct. Finally, if the appearance
of a person stands out amongst the otherwise undistinctive crowd,
then an eyewitness is more likely to select that person,
(35:51):
regardless of their own recollection of the criminal. This is
known as the distractor or dud effect. So there are
many places where scientific study has made d contributions to
the optimal way to run. I witnessed identification procedures. Now
what if somebody comes in and they say, I saw
the guy, and I know with one hundred percent certainty
(36:12):
that was the guy, Versus someone else who comes in
and says, I don't know, I was looking out a window.
It was a car robbery going on two stories below me.
It was kind of dark. I couldn't really see him
very well. I think that was the guy. I don't
really know. Do you think those two scenarios should be
treated differently? In other words, what is the relationship between
confidence and accuracy? Well? The United States Supreme Court had
(36:36):
to answer this question in nineteen seventy two in a
case called Neil versus Biggers. What happened was that the victim,
Margaret Beemer, was dragged into the woods and assaulted. She
described her attacker to the police only in very general terms.
Then she was shown a bunch of photographs and people
who met the description, but she couldn't identify anyone from
(36:57):
the photos or stand ups for seven months. So the
police then get a guy on another charge, a got
named Archie Biggers, and they decide to include him in
a stand up, But they couldn't find anyone who made
a good match to this guy's looks, so they did
what's called a show up instead, where there's no one
else there but the one suspect. So two detectives walk
(37:20):
Biggers passed the victim, and at her request, the police
directed him to say shut up or I will kill you.
The testimony a trial wasn't clear as to whether she
had first identified him and then asked that he repeat
those words, or made her identification after he'd spoken. In
any event, the victim testified that she had no doubt
(37:42):
about her identification, and so Biggers was convicted, and he
came back on appeal and argued it simply wasn't fair,
this was unreliable. The lower courts agreed with him and said,
no way, We're not accepting this as evidence because this
identification process was extremely suggestive. But the US Supreme Court
(38:03):
eventually heard this case and reversed that decision. They concluded
it was okay why because she said that she was certain.
In other words, they concluded that how confident a person
is relates to how good their evidence is. They felt
that high confidence allows you to judge the veracity of
(38:24):
a witness. So, in other words, the Supreme Court said
implicitly that they think there is a relationship between confidence
and accuracy. But generally, cognitive psychologists find that this relationship
is very weak. Because of all the things we've talked
about in this episode. Confidence does not work as a
(38:45):
yardstick of accuracy, especially as time goes on and I'm
going to do an episode soon about how the legal
system decides which technologies to allow into their courtrooms, But
for now, I'm just going to point out that it
requires passing high scientific standards. So how do you think
eyewitness testimony compares well? A version of this question hit
(39:09):
the US Supreme Court in twenty eleven Perry versus New Hampshire.
I spoke with Sanjay Gupta on CNN about this case
when it was getting decided. The question the court was
asking was essentially, if there's lousy eyewitness testimony in a case,
should it be allowed? When does that violate the right
to do process? You'll write to a fair trial. So
(39:33):
Perry was a young man in New Hampshire who got
convicted of stealing a car, and he was convicted based
in large part on a woman who was looking out
a second story window in the dark. She couldn't see
well from that distance. So Perry got convicted in a
New Hampshire court, and they came back and tried to appeal,
and he said, how can you possibly take this woman's
(39:54):
word for it? How could you send me to jail
based on such unreliable witness testimony? The lower courts upheld
the decision, and so this question spun up to the
US Supreme Court. The question that was on the table
is does unreliable eyewitness testimony violate your constitutional rights? And
(40:15):
what the court said is, look, the only thing we
care about is whether there was suggestibility by the police.
If the police manipulated the process, then the due process
clause of the Constitution is applicable. But they said, this
does not hold generally for eyewinness testimony. Using unreliable evidence
(40:35):
against you is not unconstitutional. There's no right that you'll
have great evidence. Instead, it's the job of the jury
to figure it out, to weigh the evidence. So that's
what the court decided. And reading between the lines, what
do you think one of the concerns of the court
was here, It's this, if you introduce a question about
(40:57):
the reliability of the witness, then you would, in theory,
require a pre trial hearing to determine if her testimony
is acceptable or not. And what does that do to
every single case running in the nation. One of these
Supreme Court justices posed this during the case. He said,
what about jailhouse testimony where someone in the jail says,
(41:19):
my cellmate told me x Y and Z. Sometimes it's unreliable,
sometimes it yields fruit. Do we require a pre trial
hearing every time because it might not be reliable? So
this is an issue that would overturn the way the
whole court system works. And so, at least for the
time being, the legal system knows that eyewitness testimony is terrible.
(41:45):
It can be very unreliable, and yet with all the caveats,
there's no choice but to let it remain. So in
closing your memory is not necessarily what you think it is.
Memory is not writing down what happened in zeros and
ones and then reading that back out. There are many
(42:05):
things that get in the way of good encoding and retrieval.
I really recommend Jennifer Thompson and Ronald Cotton's book. For Jennifer,
this was a complete pouring out of her guts on
the table a cathartic enterprise where she had to allow
that she put a man in jail for eleven years
based on her certainty that that was the face she
(42:27):
had looked into. She knew it was Ronald Cotton, and
she wanted him to die. That's how she felt about
this man. And then she found out that she was wrong.
So the book is about what it's like from the
inside to believe, you know, and all the steps that
happened along the way that manipulated her memory. So to
summarize what we talked about today, eyewitness testimony has a
(42:49):
terrific sway on jurors, and yet it is variable in
its accuracy. Now, just for clarity, it's not that eyewitness
testimony always has to be wrong. Plenty of times someone
IDs someone and it's correct. But memory is not like
a video camera. It is a reconstruction, and many factors
worse than the encoding and manipulate the retrieval. As a result,
(43:13):
we cannot say that confidence and accuracy are tightly linked,
especially as time goes on. Eyewitness testimony is not going
away from the courtroom because often it's the only evidence
that we can bring to bear in a case. So,
even though it may be the worst technology that we
allow in the courtrooms, it is presumably here to stay.
(43:35):
But remember, like much about our perception of the world
that I'll be discussing throughout these episodes, just because you
believe something to be true doesn't necessitate that it is.
Go to Eagleman dot Com, Slash podcast for more information
(43:56):
and to find further readings and to see some photographs.
Send me in email at podcasts at eagleman dot com
with questions or discussion, and I'll be making an episode
soon in which I address those. Until next time, I'm
Dave Eagleman, and this is Inner Cosmos
Host
David Eagleman
Popular Podcasts
40s and Free Agents: NFL Draft Season
Daniel Jeremiah of Move the Sticks and Gregg Rosenthal of NFL Daily join forces to break down every team's needs this offseason.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.