Investigators botched their interviews of hundreds of children in the 1980s McMartin Preschool abuse case, critics charge.
Two decades later, "it’s still questioned back and forth, about whether anything happened at McMartin or not," said UCLA psychology professor R. Edward Geiselman. An internationally recognized expert on how such interviews of children as well as adults should be conducted, Geiselman is among the many critics who say the case was botched by investigators who asked the kids leading and coercive questions that produced evidence that couldn’t hold up in court.
"If you go back in the transcripts, things were said like, ‘Mary’s got a good memory. Don’t you want to have a good memory?’" to get children to verify each other’s testimony, Geiselman said. "There is some evidence that gives a reason to believe some things happened. But that’s all for naught because of the way the interviews were conducted." The case, he said, "was a watershed moment. The world of interviewing and how child abuse cases are handled changed across the country," and Geiselman played a big part in instituting those changes.
Psychology professor G. Edward Geiselman was contracted by the U.S. Department of Justice to develop a new approach to interviewing children.
In the early 1990s, reports of child abuse and neglect skyrocketed to such an extent that the U.S. Department of Justice (DOJ) saw such crimes against children as the nation’s No. 1 social problem.
Geiselman and UCLA psychology colleague Robert Fisher — who is now at Florida International University — were contracted by the DOJ to develop an interview technique for children that would build on their earlier work; In the 1980s, based on their research, they developed the cognitive interview for adult victims and witnesses of crime. Geiselman and Fisher developed the cognitive interview for children in consultation with the L.A. County Sheriff’s and Los Angeles Police departments and investigators in social services and law enforcement throughout Southern California.
Today, the cognitive interview, compared to the previously used Q&A format, is being used effectively by law enforcement around the world to elicit more accurate information by building rapport, letting the interviewee do most of the talking and using memory-prompting techniques.
Geiselman has taught it to thousands of psychologists and detectives who specialize in interviewing children, and the interview has been credited with aiding in the investigation and resolution of countless child abuse cases. In April, L.A. County Sheriff Lee Baca presented Geiselman with the Mary Ellen McCormack Award for lifetime contributions to the prevention and investigation of child abuse. The award is named for a 9-year-old child who was the victim of abuse in 1874 in New York. That crime led to founding of the New York Society for the Prevention of Cruelty to Children, believed to be the first child protective agency in the world.
The cognitive interview factors in developmental differences between adults and children — factors such as the degree of discomfort a person feels in unfamiliar situations.
"[Adult] witnesses walk into a law enforcement interview feeling somewhat uncomfortable or apprehensive," said Geiselman. Children feel that more profoundly. "They’re busy worrying about how they’re supposed to behave in this situation. You need to calm them down and develop rapport so that they’re able to focus all their cognitive resources on remembering things." Once there’s rapport, a child will be invited to reconstruct the event. "It’s like going back to the scene of the crime, but in their head," said Geiselman. "If we have an allegation that Uncle Jimmy molested the child sometime last night, I would ask, ‘Do you remember when Uncle Jimmy came over last night? What were you doing right before Uncle Jimmy came over?’ I try to get them back there. Sights, sounds, smells … You get them to slow down their thinking … The memory’s going to be more complete."
Building rapport is the first step in Geiselman's approach to interviewing children who may have been victims of abuse.
As the child tells the story of what happened — however he wants to tell it and however long it takes — he or she "tells the story the way their memory is laid out," said Geiselman. This way, "I’ll get more information, and I’ll get more accurate information."
The interviewer’s job is to just listen. "You don’t interrupt them — I don’t care how much they’re stumbling around," the psychologist said. "You just listen."
Giving a detailed narrative is difficult for children, especially under the age of 7. And children also feel understandably embarrassed or afraid of talking about sexual abuse and similar incidents. So when the child is finished, Geiselman advised, the interviewer can get more details by asking open-ended questions, like "You mentioned that Uncle Jimmy played games with you. What were the games like?"
Interviewers can also use memory-jogging tools to elicit additional information. One L.A. County detective got results by changing the child’s perspective when the boy became uncomfortable talking about being molested in his room. "The detective knew that the child had a California raisin (stuffed toy) up on a shelf in his room. Near the end of the interview, he asked, ‘When so-and-so came into your room, what would the California raisin have seen?’ And the child told him a lot of stuff that that led to this particular person, I believe, getting a 562-year sentence."
Investigators also avoid asking the kinds of leading questions that contaminated the McMartin case. "Anytime you insert information into a question that the interviewee has not told you, that’s a leading question," Geiselman said. For example, "an interviewer might say, ‘Now when Uncle Jimmy touched you’ …. and there hadn’t been any discussion of touching.
"The goal of the cognitive interview is to maximize the amount of information gained while preserving the rate of accuracy of that information," he said. "We don’t want to screw things up on our way to generating more information," as in the McMartin case.
"The motivations for questioning children in a particular way may be real admirable, but you can really mess it up so that you can’t get convictions in court," Geiselman said.