Scientists have decades of research showing that our memories for events are sketchy—even the powerful-feeling flashbulb sort. But the annals of law courts (and advocate websites) continue to carry reports of false identifications leading to wrongful convictions and unearned prison terms. Why?
It’s not because judges don’t see it. “The mind is not a video camera,” retired Arizona superior court judge Ronald Reinstein told us during the recent annual meeting of the American Association for the Advancement of Science (AAAS). "Memory is very fragile; it can deteriorate, it can be corrupted—and we see this every day in the court."
“In the last couple of years, the courts are talking about [problems with eyewitness testimony] more and more, but most still have not caught up,” he said. One problem is accuracy of the original memory: for example, if the perpetrator showed a gun during the crime, witnesses often miss or misinterpret other parts of the scene. Another problem he described is the many variables involved in asking a witness to choose someone from a police lineup. For example, the people chosen as “fillers,” filling out a physical lineup or included in a photo spread, can influence how likely a witness will be to make a mistake. He showed us a photo of one lineup where only one man, a filler, was wearing shorts. He was chosen, partly because he stood out as different, and partly because at the time of the crime, the perpetrator was wearing shorts.
The chance that witnesses will ID the wrong person increases when they are shown a group of photos (“simultaneous lineup”) rather than one at a time (“sequential lineup”), according to the initial results of a field study done by the American Judicature Society. Witnesses identified a filler as the culprit 18 percent of the time when they were shown a simultaneous lineup, and only 12 percent of the time when shown a sequential one. Other influences on witnesses are what exact instructions they are given, whether they are asked how confident they are about the ID, and who is administering the lineup (the investigator or a neutral person).
“We’re not saying police are trying to cue the witness, but [advising them] to avoid unintentional influences,” Reinstein said. Some reforms this study and others suggest are mandatory recording of the ID procedure (video or audio), better instructions, better selection of fillers, and double-blind presentation.
One court that has caught up is that of New Jersey; Reinstein cited the 2011 case State v. Henderson, which updated the state's test for evaluating and admitting eyewitness identifications in criminal proceedings (here’s a summary of the case by the Library of Congress and the detailed report of the special master’s investigation). AAAS itself offers judicial seminars on emerging issues in neuroscience, including memory and the limits of brain-scanning technology (a Dana Foundation grant helps support this).
As for juries, Judge Reinstein offered a bit of good news: It appears that “the CSI effect” may not exist. That’s the idea that watching fictional sleuths use magical technology to solve every crime beyond the shadow of a doubt could have lured jurors into thinking that they should demand that level of technology–and surety—in a real courtroom. “I don’t believe there’s a true CSI effect,” Reinstein said.