Monday, August 11, 2008

Barry Scheck on eyewitness ID best practices

In its coverage of last week's initial meeting of the Court of Criminal Appeals' Criminal Justice Integrity Unit, Mary Alice Robbins at Texas Lawyer described a presentation by national Innocence Project co-director Barry Scheck regarding best practices for eyewitness ID procedures:
The meeting also featured a presentation by Barry Scheck, co-director of New York's Innocence Project, on the use of a computerized photo lineup program. With the high-tech software program, photo lineups can be downloaded onto a laptop computer that law enforcement can take to a witness in the field, Scheck said.

Scheck said the computer makes an audio or video recording as the witness goes through the photo array so there will be some form of documentation of the identification procedure. Also in the computer is information about the witness, he said. That includes the witness' name, his or her description of the suspect, the distance the witness was from the suspect at the time the crime was committed, and whether the witness was under the influence of drugs and alcohol.

For those law enforcement agencies that prefer less high-tech procedures, Scheck suggested the folder method. With that method, the person administering a photographic lineup places each photo in a separate folder, Scheck said. The witness then looks at the photo in each folder separately.
The discussion above referenced Scheck's advocacy of "blind administration" of lineups, including his low-tech "folder method" for performing a random, blind photo lineup. Some agencies have said using blind administrators could add onerous staffing demands, but Scheck said that using either a laptop or the "folder method" could meet the blind administration requirement without additional staff. (See the national Innocence Project's web page on eyewitness misidentification.)

Scheck also told the CJIU about several best practices that are cost free and merely need to be mandated in the law. Chief among those is telling the witness that the perpetrator may or may not be in the lineup. This reduces error, he said, without reducing accurate IDs and costs nothing. Similarly, witnesses should be asked to describe their level of certainty in the identification in their own words, and officers should avoid admonitions to witnesses like, "you have to be sure." Scheck also encouraged audio or preferably video recording of lineups.

Another key, free best practice: "Fillers" chosen for the lineup should be matched to the original witness description instead of finding fillers who look like the suspect. For example, if a witness said the suspect had blue eyes but the person police pick up has green eyes, only blue-eyed fillers should be used for comparison. (See comments for the correction.)

Most of these reforms are free and do not reduce the rate of accurate IDs, said Scheck.

Scheck also advocated using "sequential" lineups instead of photo arrays, but acknowledged that more research needs to be done to definitively prove which is better. Recent research questions whether "sequential lineup advantage is dependent on lineup composition and suspect position," while other researchers insist it's the best way to go. More significant research projects currently underway may soon answer the question definitively, he said.

Related Grits coverage: New CCA "integrity unit" shows shift in establishment opinion about innocence reforms.


Anonymous said...

if a witness said the suspect had blue eyes but the person police pick up has green eyes, only blue-eyed fillers should be used for comparison.

This seems like a bad idea coming from a person who is aggressively advocating the position that eyewitnesses get some of the details wrong.

Gritsforbreakfast said...

Why, exactly, do you think it's a bad idea? It seems like the concept that the fillers should match the description, not the suspect, is pretty important to avoid IDs becoming self-fulfilling prophecies. The hope is that when witnesses get it wrong, you CATCH the error, not reinforce it.

That said, to be fair, the example about eye color was my own characterization, not Scheck's. Looking at their website, the example may contradict their admonition that "The suspect should not stand out (for example, he should not be the only member of his race in the lineup)."

I'm not sure offhand what is the right lineup composition when a suspect doesn't match the original description and therefore may stand out from the fillers. It's a good question. I'll try to find some clarification.

Soronel Haetir said...

I too am interested in what the correct answer is for the case where the suspect does not match the description in some manner that would make the suspect photo stand out.

Gritsforbreakfast said...

Okay, I think I cleared it up and confirmed my original interpretation of Scheck's position on this point was in error. A policy researcher who asked not to be named offered this reply:

"In the hypothetical you offer, the fillers should have the same eye color as the suspect. You should still make all of the lineup members match the description given by the eyewitness. However, lineup members should all have the unique characteristics of the suspect.

"You can check out page 29 of the NIJ guidelines (attached), which state the following:

'Select fillers who generally fit the witness’ description of the perpetrator. When there is a limited/inadequate description of the perpetrator provided by the witness, or when the description of the perpetrator differs significantly from the appearance of the suspect, fillers should resemble the suspect in significant features.'

"I believe this was empirically tested in the following article as well: Wells, G.L., S.M. Rydell, and E.P. Seelau. “On the Selection of Distractors for Eyewitness Lineups.” Journal of Applied Psychology 78 (1993): 835–844."

So the literature recognizes there's a tension between these two requirements and the admonition that the suspect shouldn't "stand out" apparently trumps. Sorry for the error - I was writing from my notes and apparently just misinterpreted.

mapi said...

I love your blog, I found a Private Investigator blog you may find useful. said...

For my part one and all have to go through this.