"For the first time in New York since the state's high court opened courtroom doors to identification experts, a judge has said a jury can hear testimony that suggests eyewitnesses are particularly unreliable when identifying perpetrators of a different ethnicity. The judge rejected prosecutors' arguments that scientific experts had not reached generally accepted theories on the accuracy of "cross-racial" IDs."

"It has become one of the most researched issues in eyewitness identification today," Justice Massaro wrote in People v. Radcliffe, 3714/01. "Inexorably, the studies are conclusive that human perception is inexact and that human memory is fallible; where cross racial-identification is involved, this is especially so." ...The judge said that although courts can give jurors a separate instruction when a criminal case relies on one witness, the question of cross-racial identification is not covered by the instruction and could require expert testimony if certain circumstances are met.

Eyewitness identification is the leading cause of wrongful convictions. Juries need to be educated about the principles of memory and proper versus improper identification techniques. This opinion is a major step in that direction.

It should be noted that the Judge refused to allow expert testimony on the issue of eyewitness confidence. We think this is in error given the current state of psychological research showing that jurors tend to overbelieve eyewitnesses and that a witness' confidence in their selection of a suspect from a photo or live lineup is actually a weak indicator of reliability.