Your lying eyes

Sam Sommers, a social psychologist at Tufts University, has a problem with photo arrays. While preparing to testify as an expert on eyewitness memory, he got a funny feeling about the photo array the complaining witness had viewed.

Mr. Sommers took that photo array, composed of nine photographs, and showed it to 31 people who matched the same basic demographics of the alleged victims. Without providing any details of the crime, he asked each subject to pick the person out of the photo array whom he or she thought committed the crime. He then asked them to make a second choice.

Basic probability tells you that the random sample should pick out the defendant about 11% of the time. Adding a second choice would increase the odds to 2 in 9. The results of the experiment astounded Mr. Sommers.

But in my photo array experiment, 23% of naïve respondents picked out the defendant with their first choice, knowing nothing at all about the crime. And a full 45% chose the defendant with either their first or second choice. Statistical analysis confirms that these are significant deviations from chance: for some reason (or reasons), the defendant did stick out like a sore thumb, casting doubt on the usefulness of the actual victims' identifications in the case.

Mr. Sommers decided to run the experiment again to see if the results would be different if he provided details of the crime. For the follow-up he provided the subjects with the same description the victims gave to police.

The second time around the test subjects picked the defendant out of the photo array 29% of the time and 65% of the test subjects picked him out with their first or second choice.

Whether this experiment speaks more to the problems with photo arrays or with eyewitness identification I don't know. It's shocking that the number of people who picked the defendant out of the array, without having any knowledge of the crime or the suspect, was more than twice what blind chance would dictate.

What was it about the photograph? Was the image the same size as the others on the page? Was it the barely visible booking identification number? Did he just have that look?

Perhaps a new method of eyewitness identification needs to be used - such as that The Justice Project proposed in March of last year.

Manufacturing memories

According to this article in the new issue of Scientific American, researchers at the University of California, Irvine, were able to manufacture memories for 25% of the subjects in a memory study. The ability to "create" memories of events that never took place places further doubt about the reliability of identification testimony by eyewitnesses.

In the study, Drs. Elizabeth Loftus and Jacqueline Pickrell gave test subjects written accounts of four incidents that took place during the subjects' lives. Three of the accounts were of actual events the occurred and the fourth was made up. Researchers made up a tale of the test subject being lost at a shopping mall at a young age. To make it more "authentic," a relative provided details about the mall. Surprisingly enough, 25% of the test subjects said they remembered the incident -- even though it never happened.

The act of remembering, says eminent memory researcher and psychologist Elizabeth F. Loftus of the University of California, Irvine, is “more akin to putting puzzle pieces together than retrieving a video recording.”

Unfortunately jurors put great weight on identification testimony, even though the eyewitness may be testifying about an event that took place years before. Jurors place even more weight on the testimony if the witness claims to be very confident about the identification (even though studies have indicated that the confidence of the witness has very little to do with the accuracy of the identification).

The barring of expert testimony regarding the fallibility of eyewitness identification lends additional credence to identification testimony, as courts seem loathe to allow testimony regarding laboratory studies on memory. Yet, according to The Innocence Project, 73% of the people exonerated as the result of DNA testing were convicted largely on eyewitness testimony.

When considering eyewitness testimony, just remember that old experiment where someone ran into the room while the teacher was lecturing. Do you recall how many people had different memories of the incident? Just imagine what would happen when you add the stress of witnessing an actual criminal event.

Death of a bill

In the most recent biennial session of the Texas Legislature, a bill commissioning an institute at Sam Houston State University (a school with one of the premier criminal justice programs in the nation) to develop best practices in both live line-up and photo-array identifications, died on the floor after a fight to defeat a voter ID bill.

The proposed legislation would have required law enforcement agencies to follow the institute's findings in developing their own written procedures governing how line-ups would be conducted.

Researchers say it is important that the officer presenting the lineup not know details of the case or who police suspect, so as not to consciously or unconsciously influence the witness.

It's also important to specifically warn the witness that the perpetrator may not be among the photos shown or among the suspects presented in a live lineup.

So why would anyone object to a law that would require law enforcement to use the best scientifically proven techniques?

In the last 15 years, 39 Texas convictions were overturned by DNA evidence. All six Harris County convictions overturned hinged upon false eyewitness identifications.

Back in March, I wrote a piece that incorporated six safeguards The Justice Project called for regarding eyewitness identifications. Those safeguards are:

• Law enforcement agencies should document lineups or photo arrays by providing the photos used in a photo array or a photograph of the live lineup as well as all dialogue and witness statements made during the procedure.
• Law enforcement agencies should inform eyewitnesses, prior to viewing a live lineup or photo array that the accused may or may not be in the lineup. Eyewitnesses should not feel compelled to make an identification.
• Lineups and photo arrays should be composed "fairly." The report calls for fillers to be selected based not on their resemblance to the accused but, instead, on their resemblence to the description provided by eyewitnesses.
• The person conducting the lineup or photo array should not know the identity of the accused. Having a "blind" person conduct the lineup reduces the chances of the officer influencing the results of the lineup or photo array.
• Law enforcement agencies should avoid exposing eyewitnesses to multiple viewings of the accused.
• Law enforcement agencies should consider using sequential arrays instead of traditional arrays. In a sequential array, an eyewitness views one individual at a time which prevents the witness from making an identification through process of elimination.

Gary Blankinship, president of the Houston Police Officer's Union, claims credit for killing the bill in Badge and Gun, the union's official publication. This claim despite the fact the bill made it through both a House committe and the (all) powerful Calendar committee before dying on the floor.

You can't always believe what you (think) you see

On Wednesday The Justice Project released a scathing report entitled "Convicting the Innocent: Texas Justice Derailed" in which TJP analyzed the reasons that thirty-nine innocent citizens spent over 500 years in prison before being exonerated by DNA evidence. According to the research, eyewitness identification "is by far the leading factor in wrong convictions in Texas." 

"Any wrongful conviction is a tragedy because it leaves the guilty unpunished and condemns the innocent to prison, or death." -- Wallace B. Jefferson, Chief Justice, Texas Supreme Court

My colleague, Houston criminal defense attorney Mark Bennett commented on my blog post about Harris County DA Pat Lykos' report on the Rachell exoneration, that eyewitness identification was a much more serious issue than faulty DNA analysis. Mr. Bennett is correct in that assessment as DNA evidence is present in but a small minority of cases.

Per TJP's report:

"While the majority of this report focuses on the wrongful convictions uncovered through DNA testing in Texas, they are only the tip of the iceberg. The advent of DNA technology has given our criminal justice system a tool that can provide incontrovertible evidence of guilt - and innocence - in cases where the presence of biological evidence is dispositive. Unfortunately, biological evidence is present in only a fraction of criminal cases. While DNA is an invaluable tool, it does not solve the problems of unreliable evidence that repeatedly surface when wrongful convictions are discovered. The vast majority of cases simply do not have probative DNA evidence."

This lack of probative DNA evidence presents a substantial hurdle to most citizens seeking exoneration - how do you prove your innocence?

"While a defendant is innocent until [should really read "unless"] the prosecution proves guilt, after a conviction occurs, the burden shifts to the defendant to prove innocence. New evidence that merely casts doubt on the conviction is not nearly enough to overturn a conviction - which is why DNA evidence, where it exists, is so successful in exonerating the innocent. Without DNA evidence, inmates face an almost insurmountable challenge to establish their innocence conclusively."

The report lists the following facts regarding exonerations in Texas:

• Texas has had more wrongful convictions exposed by DNA (39) than any other state in the country;
  
• The 39 exonerated citizens spent over 548 years in prison (an average of 14 years each);
  
• State and local governments have paid out over $17 million in compensation to the exonerated;
  
• Twelve Texas counties have uncovered wrongful convictions through DNA evidence;
  
• Dallas County leads in the number of wrongful convictions because Dallas County preserves DNA evidence while other counties destroy it;
  
• Nine citizens have been freed from death row based on evidence of innocence;
  
• 85% of the wrongful convictions involved eyewitness identification;
  
• 28% of the wrongful convictions involved the use of unreliable or limited forensic methodologies;
  
• 18% of the wrongful convictions involved false forensic testimony;
  
• 18% of the wrongful convictions involved the withholding of exculpatory evidence or other prosecutorial misconduct;
  
• 13% of the wrongful convictions involved accomplice testimony; and
  
• 13% of the wrongful convictions involved false confessions or guilty pleas.
  

While I will come back to this report in a future posting, today I am focusing on the reliability of eyewitness identification and what should be done to lessen the impact of mistaken eyewitness identification in the future.

The report points out that Texas has no statutory rules or standards regarding how live lineups or photo arrays are conducted. More troubling is the fact that some 88% of law enforcement agencies have no written policies regarding lineups and photo arrays.

TJP calls on Texas to adopt the following safeguards when conducting lineups and photo arrays:

• Law enforcement agencies should document lineups or photo arrays by providing the photos used in a photo array or a photograph of the live lineup as well as all dialogue and witness statements made during the procedure.
• Law enforcement agencies should inform eyewitnesses, prior to viewing a live lineup or photo array that the accused may or may not be in the lineup. Eyewitnesses should not feel compelled to make an identification.
• Lineups and photo arrays should be composed "fairly." The report calls for fillers to be selected based not on their resemblance to the accused but, instead, on their resemblence to the description provided by eyewitnesses.
• The person conducting the lineup or photo array should not know the identity of the accused. Having a "blind" person conduct the lineup reduces the chances of the officer influencing the results of the lineup or photo array.
• Law enforcement agencies should avoid exposing eyewitnesses to multiple viewings of the accused.
• Law enforcement agencies should consider using sequential arrays instead of traditional arrays. In a sequential array, an eyewitness views one individual at a time which prevents the witness from making an identification through process of elimination.
  

While juries seem to accept eyewitness testimony as the most accurate evidence, we know instinctively that such testimony is fraught with problems: lighting, time, memory, excitement, fear, anger and ethno-centrism among others. The proposals in The Justice Project's report are a meaningful first step in reducing the number of wrongly convicted citizens behind bars.