The Connecticut Supreme Court in December considered several cases related to eyewitness testimony, including one in which the Connecticut Psychological Association is asking the court to reform its standards for reviewing eyewitness identifications.
The CPA filed an amicus brief along with the Connecticut Criminal Defense Lawyers Association (CCDLA) in the case of the State of Connecticut v. Dashawn Revels, asking the court to revisit the criteria it considers to determine reliability of eyewitness testimony, when an identification procedure is shown to be unnecessarily suggestive. At the very least, the parties are asking the court to end the use of what it deemed the most problematic factor in the reliability analysis, “the level of certainty demonstrated at the identification.”
This action is the first time the CPA has filed an amicus brief. Traci Cipriano, JD, Ph.D., licensed clinical psychologist and CPA director of professional affairs, says research has shown that eyewitness confidence is unreliable as a predictor of accuracy and the CPA is concerned that the standards the court is using relevant to the question of the reliability of eyewitness testimony are not rooted in current psychological science. In the brief, the CPA and CCDLA express concern about bringing the state’s procedures for evaluating challenges to eyewitness testimony into accord with scientific research.
“In particular, when you look at the research, confidence is very weakly correlated with actual accuracy,” Cipriano says. “This is compounded when police use a suggestive procedure in having the suspect identified.” A suggestive procedure might include the police presenting just one suspect for identification, rather than using a lineup with multiple individuals. “Eyewitness accounts can be very vague,” she says. “So when you have a lineup, the person really has to pick among a group of individuals whereas if you just have one person, the eyewitness might just say ‘yes, that person fits the description.’”
“Suggestibility further decreases accuracy,” she says. “Even though the person feels confident – that doesn’t mean they are accurate. We really wanted the court to see that these standards should be modified.”
The Innocence Project filed a companion brief in the same case, asking the court to establish a new framework for evaluating eyewitness testimony. As of 2012, there were nearly 300 DNA exoneration cases listed on the Innocence Project Web site – and eyewitness misidentifications played a role in nearly 75 percent of them.
The CPA filed its brief right after the court’s August 2012 ruling (state vs. Guilbert) which changed the law to allow attorneys to bring in psychological experts to testify on issues related to eyewitness accuracy. “In the past, experts were not allowed to testify,” Cipriano says. “So that’s a huge change. Now there is room for the science to be brought in the courtroom.”
While Cipriano says that ruling is a step in the right direction, she says the CPA did not withdraw its amicus brief because the association would like for the courts to reconsider the eyewitness confidence prong in determining reliability. “Eyewitness confidence is so loosely correlated with accuracy that we don’t think that should be one of the criteria the court uses to examine eyewitness reliability,” she says. “We are less concerned about it than we were when we originally filed knowing now there can be expert testimony, but on the other hand, many of these defendants probably can’t afford to hire an expert in all of these cases, so ideally eliminating that criteria would be the best.”
The court heard two additional cases related to eyewitness testimony on that same day in December, including the state of Connecticut vs. Troy Artis, in which the American Psychological Association filed an amicus brief providing the Court with an overview of the strong body of research showing the variables that affect accuracy of eyewitness identification, specifically addressing the point that suggestive circumstances that will affect eyewitness identification can occur without police action and that limiting due process protections to only those faulty eyewitness identification procedures that are caused by state actors is too narrow a band of protection.