Getting the Truth from Child Witnesses

Getting the Truth from Child Witnesses image‌Cases of child abuse often turn on the eyewitness testimony of children. Sometimes that testimony is unreliable—not because children lie but because they misremember events or their memory has been shaped by questioning meant to elicit the truth.

Montclair State University researchers are in the middle of a three-year, National Science Foundation-funded project that could help investigators better interview children and understand how the use of body diagrams in forensic interviews influences the accuracy of children’s reports.

Perhaps the most famous example of how questioning can change memory is the 1980s McMartin preschool case, in which, after seven years of trials without a conviction, the accusations of dozens of children were determined to be false memories created during repetitive interviews with suggestive interviewing techniques.

“The information we’re gathering will influence state- and national-level interviewing practices as well as the content of training for professionals who investigate crimes against children and other vulnerable groups,” says Associate Professor of Psychology Jason Dickinson, who is leading the Montclair State team in a collaborative study with a research team at Central Michigan University.

The study, “The Impact of Disclosure History and Interviewing Protocol on Children’s Eyewitness Testimony,” builds on his team’s past research on the subject and uses a new paradigm that allows researchers to safely and ethically test disclosure history with “innocuous touches” and varied interviewing techniques in a laboratory setting.

The team uses an undergraduate assistant named “Mr. Science, Germ Detective,” who, while conducting germ education activities, twice touches the children in innocuous ways such as shaking hands. Prior to the session, children are told that Mr. Science can’t touch them because he might spread germs. He uses “Glo Germ,” that glows under a black light to illustrate how germs spread.

A week later, before the children are interviewed, researchers ask parents whether their children mentioned the touching. Children who did are considered “disclosers;” those who didn’t are “non-disclosers.”

 “There may be something fundamentally different between these groups in terms of how they remember events,” Dickinson explains. “For example, it is possible that because disclosers have previously reported touching they are more resistant to suggestion in a subsequent interview.”

Using this paradigm, Dickinson’s team is testing how body diagrams—two dimensional human outline drawings—affect children’s reports of touching. His previous research has found that diagrams elicit more false reports of touching. Now they are studying whether using diagrams at the end of the interview minimizes false accounts.

“To look at the big picture, we are trying to find the best ways to question children about events they have experienced. We are finding ways to maximize the accuracy of their reports and minimize errors to help the justice system identify the guilty and exonerate the innocent.”