Live from TEDGlobal

The fiction of memory: Elizabeth Loftus at TEDGlobal 2013

TG2013_027570_DSC_1945

Photo: James Duncan Davidson

Elizabeth Loftus begins her talk at TEDGlobal 2013 with the tragic story of Steve Titus, who was arrested in 1980 because he sort of matched the physical description of, and drove a similar car to, a man who had raped a woman in his area. Looking at a photo lineup, the victim told police that Titus looked "the closest" to the man who had raped her. But by the time the trial began, the victim had become certain that Titus was the attacker.

Memory scholar Elizabeth Loftus worked on this case, and became fascinated by the question: How did the victim's memory go from uncertain to certain? The stakes of this shift were unbelievably high. While Titus was eventually exonerated by a journalist who tracked down the real rapist, he lost faith in the justice system, lost his job and fianceé, and became obsessed with what had happened to him. He died of a stress-related heart attack at age 35.

Loftus studies memory, but in an unconventional way. "I don't study when people forget," she says. "I study when people remember things that didn't happen. I study false memories." It's an issue that comes up often in courts, which highly value witness testimony. In a survey of 300 cases of wrongful conviction, where a person was later exonerated of a crime, three-quarters of them had been incarcerated due to faulty human memory.

"Many people believe that memory works like recording device," says Loftus. "But decades of research has shown that's not the case. Memory is constructed and reconstructed. It's more like a Wikipedia page — you can go change it, but so can other people."

TG2013_028208_D41_9891

Photo: James Duncan Davidson

In one study conducted by Loftus, people were shown a simulated accident and were asked how fast a car was going when it either "hit" or "smashed into" a second vehicle. When the word "smashed" was used, people overestimated how fast the car was going — and reported seeing broken glass at the accident site. More recently, Loftus did a study on members of the military as they were being trained on how to handle interrogations in the event of capture. These subjects underwent a stressful physical interrogation and were later asked to identify the interrogator. If they were fed suggestive info, they would misidentify the interrogator — even if there was no physical resemblance.

"When you feed people misinformation about some experience they had, you can distort or contaminate or change their memory," says Loftus. It doesn't have to be done with malintent — it can be inadvertent.

In the early 1990s, Loftus became interested in a kind of story she was seeing more and more in the media. A psychotherapy patient would remember a brutal memory that they had repressed from childhood. When Loftus looked into some cases — including one where a woman had remembered being forced into a satanic ritual, forcibly impregnated, and having her stomach cut — she was struck that the physical evidence just didn't match up with the memory. She wondered: Were the exercises of psychotherapy actually leading to false memories? She designed a study where suggestion was used to plant a memory of being lost in a mall as a child — and it stuck with a quarter of subjects. Loftus notes that other studies conducted around the globe showed that it was easy to plant traumatic childhood memories, even for highly unlikely events.

As Loftus spoke out on what might be happening in these cases, she says, she was hit with a defamation suit from one anonymous subject. "I became part of a disturbing trend in America where scientists are being sued for speaking out on matters of controversy," she says.

TG2013_027581_DSC_1956

Photo: James Duncan Davidson

More recently, Loftus drew public fire when she presented the results of a study where a memory was planted that subjects had gotten sick after eating a certain food as a child. When later presented with the food at a picnic, they ate much less of it. The inverse worked for good memories about healthy foods. This made Loftus wonder: Could the malleability of memory be used to help establish healthier behaviors? This, of course, brings up ethical dilemmas.

"When should we use this mind technology? And should we ever ban its use?" she asks. "False memories aren't necessarily bad or unpleasant. By planting a warm fuzzy memory about a healthy food, like asparagus, we can get people to get asparagus more."

To a big laugh, she wonders if using these tools on children was any more ethically suspect than encouraging them to believe in Santa Claus. She, however, understands why the topic makes people uncomfortable.

"Most people cherish their memories," says Loftus. "But I know from my work just how much fiction is already in there."

Elizabeth Loftus's talk is now available for viewing. Watch it on TED.com»