By Elizabeth Svoboda, published on March 1, 2004 - last reviewed on November 20, 2015
Remembering Satan, Lawrence Wright's widely read book, profiles a 1980s father who "remembers" inflicting ritual abuse on his daughters. The book blames false memories primarily on interrogators who use techniques like hypnosis and leading questions.
New neuroimaging research at Johns Hopkins University, however, suggests that what happens in the brain at the moment a memory forms is just as essential to false-memory development as are retrieval methods that are used much later.
Yoko Okado, a psychology graduate student at Johns Hopkins, and her adviser, Craig Stark, wanted to find out if differences in brain activation during an event influenced subsequent memories. Their subjects first sat in a functional magnetic resonance imaging (fMRI) scanner and read short descriptions of ordinary events, each containing 12 critical details, such as "left-handed robber."
The subjects then returned to the fMRI scanner and read similar descriptions in which the 12 details had been changed (the "misinformation" phase). The participants who successfully identified which details had appeared during the first phase showed more neural activity in the hippocampus, a brain area important in establishing memories, during that initial phase.
However, participants who incorrectly identified facts from the misinformation phase as original details showed higher levels of brain activation during the misinformation phase than during the first reading of the descriptions. The results appear in The Journal of Neuroscience.
"The idea that our memories are fallible holds up here," Stark explains. "We see that some things are very well encoded by the brain, and some are not. If bits of misinformation are well-processed, that may be what causes you to say they are true." He hopes these results will provide neurological grounding for the work of Elizabeth Loftus and others who have argued that exposure to misinformation can cause memory distortion.
Stark says false memories have a lot to do with the way our brains "average away" the details of any specific episode. He says, "With a lot of memories, you can get it down to 'Oh, I learned that in French class in 10th grade.' But what day? What were you wearing? You don't know." We often latch on to a false detail that would make sense in the situation, he adds.
Stark hopes that future research will determine whether one memory can completely overlap another. "Could a strong encoding of a memory of a screwdriver be overridden by a strong encoding of a wrench? Are both of them still in your memory, or has the new one really replaced the old one?"