The Research Design Associates Blog

(eyewitness confidence)

Why Do So Many Eyewitnesses Get It Wrong?

Over 260 people have been exonerated through DNA testing in the U.S. More than 75% of these innocent people had one or more eyewitnesses misidentify them. In about 40% of these misidentified cases, multiple eyewitnesses identified the same innocent person. How can so many eyewitnesses be wrong? Many of these misidentifications could have been prevented and many wrongful convictions averted if police had used more reliable lineup procedures. Detecting problems with a lineup, that is, assessing the fairness of the lineup, can easily be done by the police before a witness views the lineup. But by using the proper precautions and testing lineups for fairness, we can decrease the likelihood of misidentifications.

Read more …


From the Lab to the Courthouse

As a psychology graduate student at Emory University in the late 1970s, I was looking for a way to apply what scientists knew about perception and memory in the laboratory to the real world. I designed a series of experiments that varied lineup size and the similarity of the decoys in the lineups to the suspect. After briefly viewing a video of a person cashing a check, subjects tried to identify that person in a lineup one week later. Based on the results of these experiments, I devised guidelines for lineups that maximized correct identifications and minimized false identifications. But I also was surprised by a counter-intuitive finding: there was no relationship between the level of confidence of the subjects and the accuracy of their identifications.

Read more …