Eye of the beholder

Study shows eyewitness testimony can be affected by inference

By Lianna Davis ’04

The brain’s apparent need to manufacture a logical sequence can trick people into believing they’ve seen something that never happened. That’s the conclusion of a study published in the Journal of Experimental Psychology: Learning, Memory and Cognition by Puget Sound Assistant Professor of Psychology Mark Reinitz and Sharon Hannigan of Boston University. The study could have far-reaching implications for courtroom justice, since it offers a scientific explanation for something people working in law enforcement have long observed: Eyewitness testimony isn’t always reliable.

Hannigan and Reinitz showed a group of undergraduates a slide show of event sequences–a woman grocery shopping, for example. In the first part of their experiment, the researchers built on previous studies to conclude that participants often said they recalled slides that had not actually been in the sequence but which were typical actions for that series, like loading groceries on the belt at the cashier line.

The confirmation that participants mistook inferences for visual memories led Hannigan and Reinitz to their second part of the experiment: causal inference. This slide sequence also showed a grocery shopping example, but Hannigan and Reinitz added a few not-so-typical slides.

"For instance, in the shopping slide sequence there was a picture of the woman picking up oranges from the floor," Reinitz said.

Hannigan and Reinitz then gave recognition tests to the students 15 minutes to 48 hours later, with some slides they had seen before and others they hadn’t.

"During the recognition test, we slipped in a slide they had never seen that was causally implied by what they had observed earlier," Reinitz said. "For instance, in the supermarket sequence there would be a picture of a woman taking an orange from the bottom of the stack. They had never seen that slide before. We found, surprisingly enough, that nearly 70 percent of people claimed very confidently to have seen that picture."

The implications of these findings could be significant, especially in the courtroom, where cases often turn on the basis of eyewitness testimony, he said.

"If a witness sees something and makes an inference, then later forgets that an inference was made, they may believe they observed the event," Reinitz explained. "If you can get people to falsely remember seeing people commit a crime, that’s a serious finding."

While others have produced false memories, Hannigan and Reinitz are the first to show that when people see an effect, their memory often manufactures the cause.

Reinitz hopes their work in inference-based memory errors will influence the legal system, although he acknowledges the widespread change he envisions will take time.

Reinitz, who taught at Boston University before coming to Puget Sound three years ago, hopes to continue his research by working with his Puget Sound colleagues to look at racial stereotypes and prejudices in false memories. He also wants to work more closely with Puget Sound undergraduates in their research studies.