The push for reform of eyewitness testimony got another exhibit entered into evidence Tuesday when Cook County Circuit Court Judge Neera Walsh ordered a new trial for Jacques Rivera. Rivera's murder conviction 21 years ago rested solely on the testimony of a then-12-year-old eyewitness who recently recanted.

Jacques Rivera
Eyewitness misidentification is the single greatest cause of wrongful convictions nationwide, according to the Innocence Project. Of 273 convictions that have been overturned because DNA showed authorities had the wrong person, eyewitnesses had pointed to the wrong person 75 percent of the time, the Innocence Project says.The percentage of wrong identifications is even higher for cases of sexual assault, according to the Center on Wrongful Convictions at Northwestern University's School of Law, which is representing Rivera, 46.
UPDATE: A study released Sept. 19 by the American Judicature Society finds witnesses who look as suspects one by one instead of a traditional all-at-once lineup are more accurate.
Recently, New Jersey's top court ordered reforms in the way eyewitness testimony is used. The court ruled that when a defendant can show some evidence that police influenced whom eyewitnesses picked out in a lineup, a pretrial hearing must be held to examine just what happened. The court also ordered that a system be developed to better explain to juries the potential flaws with eyewitness identification
In Tuesday's Cook County case, prosecutors challenged the recantation but Judge Walsh, a former assistant Cook County state's attorney, wrote: "The court finds that [Orlando] Lopez was credible in his recantation, and believes that Lopez testified erroneously when he identified petitioner as the man who shot Felix Valentin on August 27, 1988."
She also wrote: "The only reason this court sees for Lopez' recantation now is a desire to correct a mistake. The court is further swayed by the absence of a showing of any possible benefit to Lopez for his recantation."
An Aug. 25 Sun-Times editorial called for these reforms in the way eyewitness testimony is used in Illinois:
-- A "double-blind" system in which the police officer conducting a lineup or photo identification session doesn't know the identity of the suspect. Double-blind tests are used in other areas, such as medical research, to prevent those conducting tests from deliberately or unconsciously tilting the results. Police need to do the same.
-- Sequential photo lineups. When police show a victim an array of photographs all at once, the victim tends to pick one, even if it only resembles the suspect. And a review conducted by a retired New Jersey Appellate Division judge found that witnesses are likely to stay committed to the first mug shot they pick, even if it is of the wrong person. Showing mug shots one by one would help protect against that.
-- Electronic recording of the entire identification procedure. Typically, victims' confidence in their identifications grows over time, especially if police or prosecutors tell witnesses there is additional evidence, such as a confession, which gives witnesses a false memory of certainty. If the procedure is recorded, any doubts expressed by a witness at the time of the identification could be taken into account later by judges or juries.
Follow BackTalk on Twitter@stbacktalk
This blog brought to you by the Sun-Times editorial board (click on names to read bios):
I'm not surprised eyewitnesses are often wrong. I sometimes have troubled recognizing people I've known for years because they've changed their hair or got new glasses.
I agree, i know many cases where eyewitnesses are wrong.