Memories of stressful events can seem indelible. Recollections of discomfort, behavior and surroundings are recorded with piercingly sharp detail, causing many a sleepless night.

But a study released December in the International Journal of Law and Psychiatry suggests that memories formed under intense stress are more malleable than initially thought. The research reveals that recent traumatic memories are susceptible to the effects of visual misinformation, even in individuals experienced with handling stress.

“People assume ‘Oh, that wouldn’t happen to me,’” said Charles A. Morgan III GRD ’96, associate clinical professor of psychiatry at the Yale School of Medicine and lead author of the study. “But [memory distortion] really does occur, and it can happen very quickly.”

The study was conducted at a military survival school during the prisoner-of-war phase, in which active duty personnel are exposed to stressful situations mimicking the experiences of wartime prisoners. Following training, students were privately interrogated about their experiences. In the first phase, they were asked to complete a questionnaire regarding the appearance of the interrogator and the interrogation chamber.

Participants in the control group received questionnaires with “nonleading” questions, or questions that did not suggest possible answers — for example, “Was there a telephone in the room?” Experimental group participants received “leading” questions — for example, “What color was the telephone?” which falsely suggests that a telephone was present in the room.

Despite the absence of a telephone, 98 percent of those who answered leading questions endorsed the presence of a telephone in the room, compared with 10 percent who answered nonleading questions. Morgan said that participants exposed to questions implying misinformation were more likely, in each category tested, to endorse false memories.

“There is an assumption by the jury and clinicians that if an event was recent and highly stressful, you will never forget it,” he added. “But the way you ask your questions might change what people have to say.”

Participants in experimental groups were also asked to select an image of their interrogator from a series of photographs that did not contain their interrogator. When shown video clips containing either weapons, familiar faces or both, and asked to determine if weapons were present, participants who had been asked leading questions were more inclined to make wrongful identifications of faces and weapons than those who were not.

Madelon Baranoski NUR ’74, an associate professor of law and psychiatry at Yale who was not involved with the study, added that participants might be more likely to believe misinformation delivered by credible sources such as psychiatric researchers. She said future research should aim to identify other factors, such as gender, that may influence the misinformation effect.

According to Morgan, future studies in this field will address why and how certain people, and not others, are more vulnerable to false memories.

The research yields crucial implications for law enforcers, who tend to assume accuracy in eyewitness testimony. Morgan emphasized the need to teach law enforcers to ask open-ended, nonleading questions in interviews to protect witnesses from being potentially misinformed.

But reform seems to be on the way.

“I think that the legal system has really woken up to the idea that eyewitness testimony can actually be a problem, given the recent number of wrongful convictions,” said Elizabeth Loftus, distinguished professor at University of California, Irvine and member of the research team. Loftus mentioned recent efforts under way in New Jersey to educate juries about the malleability of eyewitness memory.

“It’s a slow process, trying to change things in the legal field,” Loftus said. “But if we can save a few innocent people from going to prison, it’s worth all of our attention.”

The study involved 861 participants and lasted for four and a half years.