Neuroscience puts memory on trial

By Emiliano Rodríguez Mega

SAN JOSE, Calif. — Jennifer Thompson’s eyes adjusted to the darkness as she heard the figure bend down to turn on the stereo. A pale blue light came off the machine and shone right up to his face. It was July 1984, and for the first time Thompson saw the face of the man who would rape her at knifepoint that night.

After testifying to the police, Thompson was shown a photo array of probable suspects. “I think this is the one,” she said and pointed at the picture of a man called Ronald Junior Cotton. A few weeks later, at the criminal trial, she claimed to be sure that he was her attacker. Cotton was found guilty, and the judge sentenced him to life in prison. Thompson tried to move on with her life until the spring of 1995, when she was informed that DNA evidence had confirmed Cotton’s innocence. “The shame was oppressive; the guilt was heavy,” she wrote in 2008. “Two years later I met Ronald Cotton to ask for forgiveness.”

Cases like this are not so rare. Of the 325 wrongful convictions documented by the non-profit organization the Innocence Project, more than three quarters of them are due to faulty human memory. Figuring out how malleable victims’ memories can be has become a pressing issue in neuroscience research.

Back when Cotton was on trial, it was hard to convince judges to allow evidence about the unreliability of eyewitness testimony, said Nita Farahany, who studies the legal implications of biosciences at Duke University. “Courts would exclude this type of evidence by saying that the neuroscience of memory was too new,” she said Feb. 13 during a symposium at the American Association for the Advancement of Science meeting in San Jose, Calif. More recently, however, scientists have been able to testify in criminal cases about some of the shortcomings of memory.

Elizabeth Loftus, a researcher at the University of California, Irvine, is especially interested in the creation of false memories. She has found that people’s memory may be malleable when exposed to misleading information after an event, causing us to incorporate and “remember” fabricated facts. “Out there in the real world, misinformation is everywhere. We get it when we talk to other witnesses, when we’re exposed to media coverage, and even when we’re interrogated in a suggestive fashion,” she explained. Plying people with suggestions about their past has allowed Loftus and her team to plant detailed false memories in human subjects. “When we test them to see what they remember about their childhood, some of them began to remember all these made-up experiences,” she said. She and her colleagues have managed to get people to remember experiences about getting lost in a shopping mall, nearly drowning and being rescued by a life-guard, witnessing demonic possessions, being attacked by a vicious animal, or having an accident at a family wedding and spilling punch all over the bride’s parents — none of which actually happened to them.

But why is our memory so unreliable? Answering this question will imply acknowledging that memory does not work like a recording device, said Felipe de Brigard, a neuroscientist at Duke University. “Fidelity is not the main purpose of memory. That’s not really what it cares about,” he said. Memory evolved to help us predict the possible outcomes of future events, not to record our past experiences and reproduce accurately on command.

Perhaps having more information about the workings — and failings — of memory will help courtrooms make informed decisions about cases like the one of Jennifer Thompson and Ronald Cotton. Because if there’s one thing scientists have learned from years of memory research, Loftus said, it’s that “just because somebody tells you something, and they say it with a lot of confidence, detail, and emotion, that doesn’t mean it really happened.”

Emiliano Rodríguez Mega studied biology at the National Autonomous University of Mexico, but switched gears to science writing. Two years ago, he and some friends created a blog called Historias Cienciacionales, where he often writes. His work has also appeared in some Mexican media outlets and digital magazines. Follow him on Twitter @mapache_rm or contact him directly at mega.fciencias@gmail.com.