We've all watched the scene wherein suspects were lined up behind a one-way mirror and an eyewitness was allowed to choose which one is the real perpetrator in movies or TV programs, but how accurate is this procedure?

Actually, DNA evidence has, more often than not, sent real offenders to jail and freed many innocent 'convicts' - prison inmates whose convictions, Psychology Matters say, hinged on eyewitness identification were later proven innocent by DNA testing.

Before the advent of DNA testing, psychologists Roy Malpass and Patricia Devine conducted an experiment to determine eyewitness accuracy in a lineup procedure.

They staged a crime during a lecture, wherein a male "vandal" (hired by the researchers) entered the hall, exchanged heated words with the lecturer, and then knocked over a rack of machines.

The audience was then asked to take part in a line up procedure and identify the vandal.

The researchers gave 2 sets of instructions to the audience wit subtle differences: one set impled that the witness had to choose among the suspects in the lineup. The other set implied that the witness did not have to make a choice.

To make their experiment more sophisticated, the researchers included the real "vandal" only half the time.

The results showed that witnesses made some accurate identifications, depending on the instructions they got.

Witnesses who thought they had to make a choice, picked the wrong suspect more often.

Those who thought they did not have to make a choice "rarely made false identification."

As for identifying the right "offender," witnesses who were not "induced" to choose were just as accurate as those who were induced to make a choice.

Malpass and Devine's study concluded that "simply telling witnesses that they don't have to choose one of the suspects in the lineup causes them to make fewer false identifications, and does not hinder witnesses' ability to make accurate identifications."

Source: Psychology Matters