Skip to content

Breaking News

  • Dennis Huner, an investigator with the Jefferson County Sheriff's Office,...

    Dennis Huner, an investigator with the Jefferson County Sheriff's Office, has been doing composite drawings for the agency since 2006. Huner will spend hours with witnesses and victims looking at photographs of noses, eyebrows, mouths and ears.

  • Huner points out different facial features that take longer to...

    Huner points out different facial features that take longer to draw.

of

Expand
Jordan Steffen of The Denver Post
PUBLISHED: | UPDATED:

A Missouri woman locked eyes with, and smiled at, the stranger minutes before he raped her, but she pointed to a different man during a lineup days later.

In Pennsylvania, a man thought he knew one of the masked, armed men who stormed into his home — DNA evidence later would prove him wrong.

A Texas man was convicted wrongfully of murdering a 7-year-old girl after three witnesses said they saw him in the park that day.

All three men spent years in prison before DNA evidence proved their innocence.

Tiny distractions can make a big difference when a witness or victim is faced with the difficult task of identifying a suspect.

Misidentification is the leading factor in wrongful convictions of people who later are exonerated by DNA evidence, according to the Innocence Project. Efforts by law enforcement to improve and standardize the practice of photo and physical lineups are falling short, according to a recent report by the National Academy of Sciences.

The nonprofit’s report lists 11 recommendations for law enforcement and the courts on how to eliminate factors that can cause someone to point to the wrong photo, and ensure that jurors understand all the things that can go wrong in making an identification. But an early review of the recommendations has drawn a mixed response from Colorado prosecutors, investigators and defense attorneys.

“This isn’t a question of corrupt witnesses, corrupt prosecutors or corrupt police officers,” said David Beller, president of the Colorado Criminal Defense Bar. “It is an issue of the inherent problems with the subconscious and the fact that the subconscious is suspect.”

Since 1989, DNA evidence has led to the exoneration of 321 wrongfully convicted people, according to the Innocence Project. In 72 percent of those cases, incorrect identifications by witnesses or victims led to the wrongful conviction.

Currently, there is no known case in Colorado in which an incorrect identification is blamed for a wrongful conviction, and attorneys say there are fewer and fewer cases in which an identification is the only evidence. But that hasn’t stopped Colorado’s Best Practice Committee — launched in April by the Colorado District Attorney’s Council and the Colorado attorney general’s office — from digging into the issue.

CDAC executive director Tom Raynes said the committee’s first project was to try to decide which method of showing photo lineups to witnesses leads to the most accurate identifications.

“The bottom line is nobody wants to convict an innocent person,” he said. “One innocent person convicted is too many.”

The committee’s research echoed that of the national report, which found it doesn’t matter if you show a witness one photo at a time or six photos at once. Instead, the national report, which was released this month, focused on factors that can alter a witness’ or victim’s ability to make a correct identification.

For example, the presence of a weapon at the scene of the crime can capture the full attention of a victim or witness, making it difficult for them to later recall the suspect’s face.

Someone with high anxiety while looking at a lineup is about four times more likely to make an incorrect identification. There is a higher rate of misidentification in cases where the victim or witness is a different race than the suspect, the report found.

Dennis Huner, an investigator with the Jefferson County Sheriff’s Office, has been completing composite drawings for the department since 2006. Huner will spend hours with witnesses and victims flipping through books filled with photos of noses, eyebrows, mouths and ears, patiently adding detail to facial features often described as “average.”

Lighting, the distance between the witness and the suspect at the time of the crime, and a variety of distractions affect what facial features a witness can recall. Huner has worked with victims and witnesses for hours, days or even months after the crime.

“Everyone has something that sticks out,” Huner said. “We each have different memories.”

The report includes five recommendations for ways law enforcement can curb unintentional cues or stressors that affect people making identifications during a photo lineup. Among the recommendations: implementing training for all officers, using a standardized set of instructions for witnesses and video-recording lineups.

Investigators also were urged to implement double-blind lineups, in which the detective showing the photos does not know who the suspect is. This could help eliminate unintentional changes in facial expressions or body gestures.

Greenwood Village Police Chief John Jackson, who is one of the board members helping to organize a November symposium for law enforcement on memory and lineups, says double-blind lineups are an exciting way to improve fairness in eyewitness identification. But he is less enthusiastic about another recommendation: having officers ask witnesses or victims how sure they are of their selection.

Jackson said that statement would be too subjective and could create later problems in an investigation or during a trial.

“You’ve either made a positive identification, or you didn’t,” Jackson said. “It didn’t count if you pointed at a picture and said, ‘I kind of think it is them.’ “

But Boulder County Assistant District Attorney Ryan Brackley said a confidence statement by a witness would improve the way eyewitness identifications are used in court.

“That initial confidence statement is something that, in all fairness, the jury should hear about,” Brackley said.

According to the report, most witnesses or victims are more confident about whom they identified during court testimony than when they initially make the identification. Brackley, however, said it may not be reasonable to require all law enforcement agencies — including the smaller, rural departments — to do a double-blind lineup in every case.

“That kind of one-size-fits-all isn’t something that can be applied here in Colorado,” Brackley said.

But Beller, president of the Colorado Criminal Defense Bar, says there is no reason not to implement all of the report’s recommendations — even if there are no known cases of convictions because of misidentifications in Colorado.

“The issue with eyewitness identification has been around for years,” Beller said. “We have looked to the courts to fix it. We’ve looked to law enforcement to fix it. Ultimately, it may require a legislative change.”

Of the four recommendations the report made for judges handling cases involving eyewitness identification, Beller said the practice of instructing juries about the flaws in identifications has the most potential to “level the playing field for a defendant.”

Asking a judge to weigh in on one piece of evidence, however, marks the top of a slippery slope, said Denver District Attorney Mitch Morrissey.

“The court should not be giving specific instructions or pointing out one piece of evidence over others,” Morrissey said.

That recommendation — and the rest of the report — is nothing new and no reason to reconsider statewide protocols, Morrissey said. The Denver DA, who is not a CDAC member, said several of the suggestions for investigators are already in place.

“It’s almost like the report of the day is the best practice,” Morrissey said. “You wait six months and there is another report that says, no, that’s not the way anymore.”

Jordan Steffen: 303-954-1794, jsteffen@denverpost.com or twitter.com/jsteffendp