Posts Tagged ‘eyewitness’

A Prosecutor Defends Eyewitness Identification

Wednesday, November 11th, 2015

It’s fairly well-established that eyewitness identification sucks, as a rule. There have been tons of scientific studies going back decades — and more are conducted all the time — on the reliability of eyewitness testimony. The studies generally conclude that we’re really bad at noticing things, remembering them accurately, and identifying faces we aren’t already familiar with. Oh, and we really really suck at it when the face is of someone of another race. Meanwhile, the field of neuroscience is reporting breakthroughs literally every month in our understanding of how the brain creates memories, stores memories, distorts memories, processes sensory perceptions, processes faces, recognizes faces, et cetera et cetera et cetera. On top of all that, there have been studies demonstrating how traditional law enforcement methods can make all of this even worse, and what other methods would work better. And on top of all that is the incontrovertible data that eyewitness testimony has played a significant role in wrongful convictions that have been proven to be wrongful.

If you want to go read the scientific literature, please do. There’s a lot. If you want a quick-and-dirty version, I’ve been covering it in my comic for — holy hell, a year now? For those preferring a more, say, jurisprudential approach, Judge Alex Kozinski wrote an excellent summary of the situation in Criminal Law 2.0, his preface to this summer’s Georgetown Law Journal Annual Review of Criminal Procedure. Click on the link to read his views.

But in American jurisprudence, these are still minority views. The courts are slow to adopt change, and have been incredibly slow to adapt the law of eyewitness identifications to the scientific facts. It’s starting to happen. And I fully expect us to reach critical mass within a generation if not sooner. But the majority of courts aren’t there yet.

And law enforcement isn’t entirely on board, either. You’d think they’d want to be on board, though — after all, who wants to lock up an innocent person and leave the real bad guy free to do it again? But inertia, the investment of ego, and confirmation bias are much more powerful than you may suspect. Especially in law enforcement. You’ve seen the mind-numbingly bad arguments cops and prosecutors have come up with to insist someone’s guilty even after the DNA proves he isn’t — or even to fight tooth and nail against the DNA analysis that would prove it. There’s too much invested in having been right.

And there’s a lot of ego, inertia, and confirmation bias invested in the sense that eyewitness identifications — they way they’ve traditionally been done — are just fine and dandy.

Which brings us to the recent publication of “The Unreliable Case Against the Reliability of Eyewitness Identifications: A Response to Judge Alex Kozinski,” by Connecticut appellate prosecutor Laurie Feldman.

Feldman’s piece is an attempt to say Kozinski — and all the others who have problems with eyewitness identifications and testimony — have it all wrong. Things aren’t as bad as everyone says, and we shouldn’t jump to make unnecessary reforms. Don’t fix what ain’t broke.

Click on the link to read the whole thing yourself. I, for one, don’t find it terribly well reasoned, but your mileage may vary. From what I understand, her points are these:

(1) There’s no scientific proof that jurors put too much store in eyewitness testimony.

(2) There’s no good science saying eyewitnesses aren’t reliable.

(3) Scholarship here is agenda-driven, which distorts the results. Peer review only makes this worse.

(4) Just because someone was exonerated, that doesn’t mean he was innocent. So how does that mean the eyewitness who fingered him was wrong?

(5) So what if 72% of DNA-based exonerations were cases involving false I.D.? How many cases, where DNA proved guilt, involved an eyewitness’s correct I.D.? What if it’s the same number?

(6) We’ve only exonerated a teeny tiny subfraction of a fraction of people convicted on eyewitness testimony. That sounds like they’re reliable, not unreliable.

(7) Courts are allowing experts to testify about all this, and they shouldn’t. This exposes the jury to political agendas, and makes the scientists advocates instead of objective scientists. Courts are moving too fast to adopt reforms like this.

(8) It’s unwise to jump on the latest social-science fad to fix what ain’t broke. There’s not a single study saying that double-blind experiments are better than ones where the examiner knows who the suspect is. And simultaneous lineups simply get better results than sequential ones.

All I can say is:

(1) If her footnotes are to be relied upon, she clearly has a lot of reading to do. I can suggest a bibliography if she wants one. This recent National Academy of Sciences report is a good start. Heck, she’s welcome to take every damn book off my bookshelf right now because I need the space for all my resources for the comic’s upcoming forays into Constitutional Law.

(2) Ditto.

(3) Perhaps she should stop reading scholarly writings and “social science” sources, and focus on the objective scientific studies out there. I encourage her to maintain a healthy skepticism of sociology and meta-analysis and studies with small sample size or weird methods — but the good news is there’s a whole lot of good stuff out there with good clean numbers and healthy p-values.

(4) DNA exonerations pretty clearly show he ain’t the guy. The eyewitness who absolutely positively no doubt about it said that’s they guy? Wrong.

(5) Not the issue. The issue isn’t how many guilty people are properly convicted. The issue is, when an innocent person does get wrongfully convicted, what’s causing that, and how can we prevent it from happening again? When 72% of your wrongful convictions that you know to be wrongful because the DNA says so were based on a false identification? That’s a very strong correlation. Combined with the fact (sorry, it’s true) that juries do put a whole heck of a lot of weight on that false I.D., and it’s safe to say we’ve got causation as well. (Whereas 100% of wrongful convictions may have taken place in a courtroom with a judge, a very high correlation indeed, but there’s zero evidence that the presence of a judge played any effect on the jury’s verdict, so we can disregard that one.)

(6) Intellectually dishonest. You’re saying “you guys haven’t disproven all these other cases yet, so let’s presume those IDs were good.” Ignoring the fact that only a small teeny tiny microscopic percentage of cases where the defendant insists upon his innocence are ever taken up by folks like the Innocence Project. You’re also comparing the number of hits in the sample size to the overall population being studied. The fact is, an obscenely high number of hits in this statistically significant sample probably translates to an obscenely high number of hits overall.

(7) Fine. You’re not allowed to use DNA experts, either… Seriously? Draw up a timeline of courts adopting real meaningful reforms here and show me how it’s moving too fast.

(8) Apparently the latest fad = decades of research, and basic Science 101 recommendations for how to conduct an experiment that have been understood for ages. None of this is rocket science. None of this is remotely eyebrow raising to anyone with a genuine science background. If you really don’t think double-blind studies are proven to be effective, that they eliminate intentional and unintentional suggestion by the examiner, then you really really shouldn’t have skipped that day in 8th-grade Science class. As for simultaneous vs. sequential, you are ALMOST RIGHT!! Because sequential arrays DO result in fewer correct identifications, along with fewer false ones. Sequential arrays only solve the problem of relative-judgment witnesses who compare the faces in the lineup to each other. They’re best used with children, lower-IQ people, and all the people whose brains really work this way only they don’t know it and neither do you. All of whom, it so happens, are the ones making the MAJORITY of false identifications. Losing a few correct IDs (because during sequential procedures we all tend to raise our threshold for certainty, worrying that a better match might be coming up) is a small price to pay for ensuring so many innocents don’t have their lives and liberty and futures taken away by mistake. Isn’t it? And simultaneous lineups are only as effective once you’ve adopted the suggested reforms, including some you didn’t discuss.

This all sounds like another one of those prosecutors thwarting justice in the name of inertia.

My final recommendation to her would to be to allow herself to be receptive to the idea that innocent people do sometimes get convicted, that bad identifications do sometimes cause this, and that there are solutions that can prevent such injustices, can ensure that the cops don’t put the wrong guy in jail, can ensure that the real perp doesn’t remain free to do it again, don’t cost hardly anything, and are easy to institute. And if that’s the case, then why the heck are you so opposed to it? Do you want those preventable injustices to occur?

Because that’s kinda how you came off.

Correct, but Wrong: SCOTUS on Unreliable Eyewitness Identification

Thursday, January 12th, 2012

In this Information Age, it is hard to grasp sometimes that everybody does not know everything. And yet it is so. It is common knowledge, for example, that dinosaur fossils are the bones of creatures that lived scores of millions of years ago, that terrorist hijackers flew planes into the World Trade Center and the Pentagon, and that eyewitness identification testimony is statistically as reliable as a ’78 Chevy. And yet there are tons of people who sincerely believe that fossils are just a few thousand years old, that the U.S. government conducted 9-11, and that an eyewitness I.D. is the be-all-and-end-all of Truth.

Actually, it’s not fair to lump the I.D. believers in with 9-11 conspiracy theorists, Genesis literalists, truthers and the like. The others are sort of fringe-y. But if you put 12 ordinary citizens in a jury box, of good intelligence and sound common sense, and the victim points dead at the defendant and says “there is no doubt in my mind, THAT is the man who raped me…” you can almost hear all twelve minds slamming shut. They’ve heard all they need to hear. So far as they’re concerned, this case is over.

This despite the fact that study after study after study reinforces the fact that eyewitness testimony sucks.

And innocent people go to jail — or worse — because of it.

So you can imagine how keen the legal world was to get the Supreme Court’s decision in Perry v. New Hampshire, which came down yesterday. Perry, identified by an eyewitness as someone she’d seen breaking into cars, argued that Due Process required a judicial hearing on the reliability of that testimony before it could be admitted at trial.

Which was the exact wrong thing to argue.

Due Process requires that the government makes sure that it does not do things that make its identification procedures unreliable. It does not require that a judge do the jury’s job. Particularly when that job — weighing the reliability of a given bit of testimony — is incredibly fact-specific.

And especially given all the evidence of all the various factors that go into making eyewitness testimony unreliable — racial differences, time lapse, focus of attention, lighting, familiarity, stress, presence of a weapon, etc. — what judge in his right mind is going to want to be the one deciding whether this particular eyewitness’s memory is good enough?

So it’s hardly any surprise that the Supreme Court balked at Perry’s Due Process argument. By a vote of 8-1 (former prosecutor Sotomayor as the lone dissenter, none better to know the power of the EW ID) the Supremes held that, unless law enforcement is alleged to have gotten the I.D. under unnecessarily suggestive circumstances, there’s no Due Process issue and certainly no reason for a pre-trial hearing on reliability.

-=-=-=-=-

No, what Perry could have argued for is either (more…)

Even Worse than Eyewitness IDs: The Police Sketch

Wednesday, June 29th, 2011

Everyone knows that eyewitness identifications are completely reliable — that is, you can count on them to be wrong.  (Everyone does know this, right?)  If the person being identified is a stranger, the chances of a correct I.D. are slim to none.  There are lots of reasons for this.  Eyewitnesses rarely have or take the time to study and memorize a person’s features.  People of one race are awful at identifying people of another race, largely because the parts of the face which differ from person to person are different from race to race — which is why people of another race often “all look alike,” because you’re looking for cues in parts of the face that don’t vary much in that other race.  And people just generally suck at remembering details consistently and accurately.

Still, sometimes an eyewitness description is all you’ve got.  And so what if the eyewitness didn’t see every detail of the face — at least they can describe the parts they did see.  Trained sketch artists take the partial descriptions provided by eyewitnesses, and using sophisticated software they can put together composite sketches that show what the bad guy probably looks like.

We’ve all seen them on the TV news, and various crime dramas would lead one to believe that they’re pretty useful.  And now with IdentiKit software, the details can be adjusted here and there until the witness goes “that’s him!”

But we never hear, after the fact, whether the drawing wound up being all that accurate.  There’s a good reason for this.  The odds of the drawing being accurate are so low, they are below statistical significance.  You’ve probably noticed this yourself, on the rare occasion when a police sketch has later been released with a photo of the culprit — the resemblance even then is usually pretty slim.

A thorough study of composite sketches by Charlie Frowd, of the University of Stirling in Scotland, had participants study a photograph of an individual for a full minute, then describe the face for a trained police sketch artist.  How well could people then recognize the faces in these sketches?  The recognition rate was as low as 3%.

Three percent.

MIT scientists Pawan Sinha, Benjamin Balas, Yuri Ortrovsky and Richard Russell have a great article here that describes problems with composite sketches and ways to make the software better.

The image above was taken from that article.  A trained and experienced IdentiKit officer was given actual photographs of celebrities with distinctly recognizable faces.  He was given all the time in the world — no pressures — and worked directly from the photos themselves instead of having to rely on another person’s descriptions.  And those sketches you see up there are the best the software could do.

Well, maybe the problem is with what the IdentiKit tries to do.  After all, it just works on individual features one at a time.  The eyes, nose, mouth, etc. are worked on in isolation.  Humans don’t look at features in isolation, though.  So there’s another kit out there called EvoFit, that’s more like a photo array that gets to evolve.  The witness is shown 72 random faces.  She picks out the six that most resemble the culprit.  The facial features of those six are then scrambled and recombined to make 72 new pictures.  The witness then picks out, again, the six who most resemble the culprit.  The process is repeated once more to get an image that pretty much matches what the witness saw in her mind.

Now, there are tons of problems with this method.  The suggestivity of showing pictures is pronounced — when witnesses choose photos from an array, they often choose not the one that closest resembles the culprit, but instead pick the one that looks different from the rest — and when a picture has been chosen, that image often replaces the image in the witness’ memory.  She now remembers that face as being the face of her attacker, even though it wasn’t.  This method of scrambling digital faces poses the same problems.

Still, it is more reliable than the IdentiKit.  Instead of a 3% recognition rate, the EvoKit attains a whopping 25% recognition rate.

One in four.

-=-=-=-=-

People suck at identifying strangers.  Period.  And yet in-court eyewitness identifications are the nuclear bombs of trial.  The victim points at the defendant and says he’s the one what done it, and you can see the jurors’ minds turning off.  So far as they’re concerned, this trial’s over.  The defense lawyer’s got a lot of work to do, now, to overcome that.

What would be just and fair, of course, would be to allow some evidence of the unreliability of eyewitness identifications in general, and the reasons why IDs can be wrong, so that the defense can tie them to specific testimony by the eyewitness to show that she made the same mistakes.  Not asking the jury to make a logical fallacy that, because it happens a lot in general, it must have happened here as well.  But actually drawing the jury’s attention to specific reasons why this particular testimony is not trustworthy, supported by expert testimony on the unreliability of IDs.

Don’t hold your breath waiting for that to happen, though.