EYEWITNESS ID BEST USING COMPUTER

A double-blind lineup on a computer sounds logical to me.

POLICE LINE-UPS ARE ‘ALL WRONG’: LOOKING AT SUSPECTS ONE BY ONE ON A COMPUTER WOULD CATCH MORE CRIMINALS, SAYS EXPERT
By Lee Moran

Mail Online
September 19, 2011

Police line-ups should be stopped and witnesses should look at suspects one-by-one on a computer, a study has found.

This would make the choice more accurate and lead to the catching of more criminals, according to research published today by the American Judicature Society.

Using the sequential method would also make it less likely that witnesses will pick the innocents brought in to to fill out the lineup, it said.

Gary Wells, eyewitness ID expert at Iowa State University, said that line-ups seen in films such as The Usual Suspects, and used in many real-life police departments, were 'all wrong'.

He found that witnesses should instead look at individuals one-by-one with a detective who also does not know the identity of the real suspect.

This, he said, was known as a double-blind lineup and avoided giving witnesses unintentional cues.

He also revealed that the 'identification parade' should preferably take place on a computer, with the theory being that witnesses using the sequential lineup will compare each person to the perpetrator in their memory.

This is in contrast to comparing them to one another side-by-side to see which most resembles the criminal.

Mr Wells said: 'What we want the witness to do is don't decide who looks most like the perpetrator, but decide whether the perpetrator is there or not.'

He said the results confirmed many other laboratory experiments, carried out over the past 35 years, that have found sequential lineups to be more accurate.

But he said some police departments have been reluctant to change their practices.

The study used real-life witnesses who did not know they were part of a study, and was conducted at four police departments in Texas, North Carolina, San Diego and Arizona.

The witnesses were shown mugshots of one suspect with five 'fillers', or known innocents.

In the traditional lineups witnesses picked a filler 18 per cent of the time, compared to only 12 per cent of the time when using the sequential method.

Witnesses picked the suspect out about a quarter of the time using both methods.

Mr Wells estimated that between 20 and 25 percent of 16,000 law enforcement agencies in the U.S. are now using the sequential and double-blind procedures.

He said those reforms have been made in the last decade, with some key departments including Denver and Dallas coming on board recently.

But he added: 'There's still a long ways to go'. And he said he hoped the study would help 'push reforms forward'.