The Duke Case

I don’t have any opinion on the guilt or innocence of Duke lacrosse players charged with raping a exotic dancer.  But I do think the incident affords a small opportunity to think about the value of eyewitness testimony.  The accuser, who is black, identified two players alleged to have assaulted her after looking over photographs of all of the white players on the lacrosse squad.   On the basis of that identification, the players were indicted by a grand jury.  So far, the accuser’s testimony is the most powerful fact in the prosecution’s arsenal.  The question is: how much credence should we give that identification?

Psychologists don’t particularly like eye witness testimony. Elizabeth Loftus has done a lot of really interesting work exposing its various frailties.   Juries and laypeople (and prosecution attorneys) tend to have a great deal more faith in someone’s ability to pick a suspect out of a lineup than they should.   In Blink, I mentioned the research of Jonathan Schooler on lineups: he’s showed that merely requiring people to write down a physical description of the suspect before viewing the lineup radical impairs their ability to pick out the correct person.

But the Duke case is an example of another, even more problematic aspect of eyewitness identifications, and that is that we aren’t particular good at making them across races.   There is a huge amount of psychological research in their area, pioneered by Roy Malpass at the University of Texas at El Paso.   A few years ago, John Brigham and Christian Meissner did a big meta-analysis of all of the cross-racial identification studies and concluded that given the task of picking someone out of a lineup, the average person is something like 1.4 times more likely to correctly identify an own-race face than a different-race face, and 1.6 times more likely to incorrectly identify a different race face.   These are not trivial error rates. Clearly we need to treat cross-racial identifications with a special level of caution.  (Here’s the link to the UTEP eyewitness laboratory:   eyewitness.utep.edu/race.html)

The problem seems to be that when we encounter someone from a different group we process them at the group level.   We code the face in our memory under the category black or white, and not under the category of someone with, say, an oval face and brown eyes and a prominent chin.   Race, in other words, trumps other visual features that would be more helpful in distinguishing one person from another.   Why do we do this?   One idea is simply that it’s a result of lack of familiarity:   that the more we “know” a racial type, the more sophisticated our encoding becomes. Another idea is that it’s a manifestation of in-group/out-group bias.   The thing about coding by group and not by facial feature is that it’s a lot faster.   And from an evolutionary standpoint, you’d want to use quicker processing methodologies in dealing with those who come from unfamiliar—and potentially unfriendly—groups.   The bottom line is that the adage that “all blacks look the same” to whites (and all whites look the same blacks) has some real foundation.

This has been a huge issue for years in white identifications of black suspects. I would venture to guess that there are thousands of African Americans in prison right now for crimes they didn’t commit, largely because whites have far too much faith in their ability to tell one black face from another.   Now, in the Duke case, we have a black identification of white suspects.   The shoe is on the other foot. It will be interesting to see whether the legal system is any more willing to acknowledge the real limitations of eye-witness identifications when it is suspects from the racial majority who are on the receiving end of the bias, not the other way around.

Comments

[charlene prince]

You've raised a really interesting point about the shoe being on the other foot. Racial bias comes into play with many cases like this, but your specific question about the legal system and what issues will be raised based on witness identification for this case is one that I will now be paying more attention to based on your post.

[Jay Howard]

I'll be just as interested to see which major media outlets address this perspective of the case, as opposed to those which will choose to stick to the more obvious, less interesting angles. Nancy Grace, I'm looking at you.

[Meg]

When I related your comments here to a friend, they said that the argument of racial bias was less valid in reverse, since caucasians have a wider array of facial characteristics.

Whether or not that's true, that seems to be a somewhat prevailing opinion amongst my contemporaries.

I'm not sure I agree, but I would digress to mention that rape victims in particular have their "eyewitness" accounts challenged more than any other group.

That's a hell of a bias.

[SS]

Perhaps race trumps more useful distinguishing features because race is typically a person's most prominent visual feature.

The difference between an oval face and a round face is very subtle when compared to the difference between black skin and white skin.

[doug]

I remember reading previously that the accuser identified the accused through photos where the lacrose players had their shirts off (bare chests). Apparently scratches/abrasion may have led the accuser to identify the two men who are currently charged.

How would this affect the data re: pure facial recognition? I'm guessing that it would make the identification more reliable if she used both facial features and scratch marks to identify the accused. On the other hand, if she was relying on scratch marks only, that would throw all the facial recognition data out the window, wouldn't it?

In any case, if true, the identification through scratch marks would certainly be a wrench into the data, so to speak.

[Jay Howard]

Meg,

Where did you hear that "rape victims in particular have their 'eyewitness' accounts challenged more than any other group"?

I'm not doubting you; I would just really like to take a look at the stats if you have them. I think it's fascinating.

[Joshua Bowlin]

In response to the first part of Meg's comment, was your friend that made the statement, "Caucasians have a wider array of facial characteristics" Caucasian himself? In my close group of friends there are three Caucasians, one Mexican, one African American, and one Vietnamese person and each of them think that their own race has more distinguishing features than everyone else's race.

[Dan Kramer]

Have their been studies on mixed race people and whether their chances of identifying somebody of white/black races would be "off-center"?

[Jay]

I think the witness identification program is deeply flawed. In this case especially since the victim was probably under intense shock, which may affect her decisions. Also police departments often couch or influence a witness towards a particular suspect who they think is the real perpetrator.

[camille]

I suppose this data about the eye witness ID is interesting in an of itself, but it seems a rather unfortunate time to bring it up. I would wager to say that the reason victim/witness identification may be flawed is because people are much more in their own heads when they are being victimised and only trying to take in information that can be useful for their survival, not really any racial reasons. White people and their likenesses are omnipresent on television and billboards and plastered on buses and throughout magazines. Everyone knows what they look like and can tell them apart. I know that most people in this country still live in fairly segregated areas, but I don't like this idea that black people live in some rural backwood Blacklandia where they are unfamiliar with white features.

I also don't think this is a case of the "shoe being on the other foot". I wish we lived in such an equal society, but alas, we don't.

[Andrea Knutson]

This fits with two personal experiences of mine one even more extreme and surprising than race.

I spent some time as one of the few white teachers at a black college. The kids frequently mixed us up.

Before that I attended a college with a 35:1 male to female ratio. Many classes only had one techcoed attending. There was a girl of Chinese background who took similar classes to Irish-American me. The guys would get us mixed up. No one ever addressed me by her name but they would forget which one of us had taken the class with them.

It isn't that they did not look carefully. Each of the fewer than 100 representatives of femininity was a prized celebrity on the campus.

[baturkey]

I wonder if the recent study in Illinois comparing the accuracy of sequential versus simultaneous lineups controlled for this.

http://www.csmonitor.com/2006/0424/p01s04-usju.html

[S. E. Sciortino]

Yes, identification across racial barriers is fraught with peril. However, in this case it needn’t be a problem. If you grant credence to the reported medical evidence, then something happened in that bathroom and there are plenty of white eyewitnesses who know who was in there. All that is needed to prevent any injustice is for a few of the players to have the moral gumption to tell the prosecution what they know.

[Overworm]

http://www.pbs.org/wgbh/pages/frontline/shows/ordinary/etc/synopsis.html
Check out this PBS Frontline story about the risk of using eyewitness testimony, especially victim identification, as the only evidence against a defendant.

Some men walked into an office and robbed and brutally beat and shot some of the employees. Eventually, police arrested suspects. One of the suspects was tried, convicted, and sentenced, even though he had an alibi and the other suspects said he wasn't involved. However, a woman who was beaten and shot during the crime (and lost an eye as a result) identified the suspect.

The trial was a joke and the police and DA's office committed many errors in the handling of the case.

After years in prison, the boy was finally released as the evidence against him was false and fabricated. To this day, the victim swears he is the man who beat and shot her.

You can watch the entire episode on-line if you wish. I warn you, this is a depressing story all around.

[WOTS]

Cross-race identification is a worthy topic for discussion, but the Duke case is hardly the best jumping off point. The vaildity of an ID generally is an issue when the eyewitness did not have a long time to view the suspect, or when the eyewitness' vision was impaired for some reason. That might be because the eyewitness was the victim of a sudden crime, or it might be because someone who witnessed the crime did so through a dirty window a hundred yards away. Cross-race ID problems come in to play because the mind has been hampered in some way from registering a detailed picture of the person.

In the Duke case, the evidence seems to be that the alleged victim was at the party for some time before she was assaulted. Particularly given her profession, she may have had ample time to recognize the players, and may even have spent time with the ones who would later assualt her.

So this is not a case where the witness plays the game of picking out the suspect in a lineup of random innocents, or in a broad photo array. In the Duke case, the alleged victim was asked to identify her attackers from a discrete group of people she encountered that evening. The question posed to her via this method was "which of these guys did it" rather than "do you see anyone her you recognize." (This might be problematic itself, but not for race reasons).

Given the limited, highly tailored group of white men she was shown, it would difficult to imagine how any non-white suspect's ID could be presumed valid (absent other factors) if not in circumstances like these. Suggesting otherwise only adds unwarranted confusion to the Duke case.

[newyorkette]

This really helps me to understand why I never recognize ANYONE. I'm multiracial, so EVERYBODY looks the same to me! Mystery solved!

And yes, without sarcasm, in spite of my artistic ability (I can draw very good portraits and caricatures) I am very very very bad with faces in social situations, have looked at my best friend in a crowd and not recognized him (thought, hmm he looks like my best friend, Juan, till he called my name).

However, a person who has been subjected to a traumatic experience at the hands of a stranger my remember a face in more detail or urgency than a person who has simply been greeted politely, and should always be taken seriously until proven wrong (if wrong s/he is).

[John Brandon]

It's all just a little to much babble for me. I think it's much simpler than that. When you see two dominos, one white and one black, what do you notice first? That one is a nine and one is a six? No, you notice that they are different colors. I don't think it's racist to say that we don't notice face shape first instead of skin color: it's just that skin color is the most obvious difference, and the thing we remember.

[Meg]

Joshua -- no, my friend was Asian.

And Jay, that's pretty much conventional wisdom amongst most criminal prosecutors and defenders -- if you are prosecuting, you cannot rely on victim testimony, and if you are defending the alleged attacker, you start by undermining the victim's character first and foremost.

I don't think that's much of a secret. And while defense on any charge usually hinges on discrediting witnesses, the process is particularly heinous and personal for rape victims.

Feel free to email me if you want links to studies and commentary on the issue. I won't spam it here.

[Coty]

I first became aware of differences in cross-race identification while an undergraduate at Duke. I was struck when one of my wife's (then, girlfriend's) hallmates--who was Japanese--stated flatly that she had a hard time telling Europeans (including American's of European descent) apart.

My first thought was, "How could she not see all the differences in hair color, eye color, etc.?" Meanwhile, I thought how similar Japanese seemed to me by comparison. Of course, the issue is not what differences exist but what differences are attended to. Paying attention to hair color is a waste of energy when surrounded by people with little variation in hair color. I remember that my wife's hallmate commented that she tended to pay attention to teeth, and that that was of less use given the extensive orthodontia here in the States.

[Michael]

Here's a .pdf with behavioral measures of the own-race recognition bias and fMRI measures of brain activity from 2001. What is clear in reading even the Introduction is that the advantage is much greater for Whites than for African-Americans. There are other references making the same point in the article, especially in the conclusion. The brain activation patterns were the same. It's a good read, Malcolm, yet it takes away from your main point, I think.

http://psychology.uchicago.edu/socpsych/eberhardt.pdf

BTW, I'll be at the APS where you are giving the Bring the Family Address, and I'm looking to hearing you.

[Denise Shungu]

For Jay Howard. Many years ago I spent three years in the Congo and was fascinated to learn that people described another black African as "red", "yellow", "chocolate" and black". We would use ruddy complexion, ivory-shinned, sallow, etc. in a similar manner.
How many of your friends can give that kind of analysis to the black people they meet? I learned to distinguish some people from different tribes from each other, but it took some work.

[Nancy Jane Moore]

I recently heard Walter Mosley say that he thought black writers could write white characters more accurately than white writers could do black characters. His theory was that blacks have been forced to pay close attention to whites for what I would call survival reasons, so they know quite a bit about whites. This seems to follow the theories in the article Coty cited above.

[Sandy]

Just responding to a couple of previous points: Various studies have found that trauma, instead of sharpening your memory, tends to distort it. People will misremember how long something took, for example, and tend to focus where their fear is focused (a weapon instead of a face, for instance). Also, amount of time spent with the other person is not necessarily an indicator of how well you'll remember them. There was a case in Texas where a woman spent 4 hours with her rapist, then identified the wrong man out of a line-up. It was cross-racial (he was black, she was white) and the wrong guy spent 16 years in jail before being exonerated by DNA.

[Overworm]

Nancy Jane, I've heard Walter Mosley say that before. I agree with it.

Black people in America, especially in the past, grew up watching white people on television all the time. It doesn't take long before one can tell the difference in facial structure between Nicole Kidman and Elizabeth Taylor. In the 70's, if white people only saw blacks on television (and not in their daily lives) they only had a handful of examples of black faces. Before the seventies, there were virtually none.

And in everyday life, it wasn't uncommon for whites to barely look at black people in passing. That's where the title of the novel "The Invisible Man" (Ellison, not Wells) came from.

[Magee]

I think Richard Dawkins had it right when he suggested that the scientific method shoulld be applied to trials/questioning/line-ups.

At the very least a line-up should be double blind. The 'victim' should only be communicating with someone who has no idea why the 'victim' is there or who any of the people are in the line up.

And then maybe comeback in a week and see if the 'victim's' choice replicates.