News Release

Northwestern researchers pinpoint how false memories are formed

Peer-Reviewed Publication

Northwestern University

EVANSTON, Ill. --- False memories are the controversial subject of hotly contested arguments about the validity of repressed memories that can surface years after a traumatic event and about the credibility of eyewitness accounts in criminal trials.

Because memories are imperfect under ordinary circumstances -- forming, storing and retrieving them, with great variations in factors influencing those processes -- it is unlikely that a one-answer-fits-all will settle those controversies soon.

But a group of researchers from various disciplines at Northwestern University literally have peered into the brain to offer new evidence on the existence of false memories and how they are formed.

Published in the journal Psychological Science, the new study used MRI technology to pinpoint how people form a memory for something that didn't actually happen.

"Our challenge was to bring people into the laboratory and set up a circumstance in which they would remember something that did not happen," said Kenneth A. Paller, professor of psychology and co-investigator of the study. (Brian Gonsalves, who was a doctoral student of Paller's and who now is a post-doctoral fellow at Stanford University, is the first author of the paper.)

"We measured brain activity in people who looked at pictures of objects or imagined other objects that we asked them to visualize. Later we asked them to discriminate what they actually saw from what they imagined," Paller said.

Extending upon considerable Northwestern research on what happens in the brain when people remember versus forget, the researchers were interested in what happens differently in the brain when false memories are produced.

"We learned that the particular parts of the brain critical for generating visual images are highly activated when people imagine images such as those we presented to our study participants," said Paller.

Many of the visual images that the subjects were asked to imagine were later misremembered as actually having been seen.

"We think parts of the brain used to actually perceive an object and to imagine an object overlap," said Paller. "Thus, a vividly imagined event can leave a memory trace in the brain that's very similar to that of an experienced event. When memories are stored for perceived or imagined objects, some of the same brain areas are involved."

Take a real life example in which a police interrogator asks if you saw a particular person at a crime scene. That induces putting that person in your imagination and possibly corrupts later questioning.

"Just the fact of looking back into your memory and thinking about whether an event happened is tantamount to imagining that event happening," Paller said. "If I ask you if something happened, you imagine it happening. Later on -- a day or a year later -- if I ask about that event, you have the tough judgment of deciding what happened and what was imagined."

It is important to know that memory is fallible, Paller said. "We know that we forget quite a bit, but we're not always in touch with the idea that our memories can sometimes can be misleading."

For this procedure of measuring brain activity, people lay down in an MRI machine as they looked at a screen with a series of words, all concrete nouns, and pictures, and they wore head phones to hear what was being said. They were instructed to generate a visual image corresponding to each object that was named. For half the words, a photographic image of the object was presented. The subjects were told to make no response to photos, but only to look at each one while waiting for the next word.

They were told to make a size judgment about the objects they were to imagine. For example, if the word was cat, they were told to imagine the cat and decide if a cat is generally bigger or smaller than a video monitor.

The memory test was administered outside the scanner and began approximately 20 minutes after the scanning. Subjects heard a randomly ordered sequence of spoken words. One-third corresponded to photos they had seen, one-third to objects they had only imagined and one-third they had neither seen nor imagined. For each word, subjects decided whether or not they had viewed a photo of the named object during the study phase.

Three brain areas (precuneus, right inferior parietal cortex and anterior cingulate) showed greater responses in the study phase to words that would later be falsely remembered as having been presented with photos, compared to words that were not later misremembered as having been presented with photos. The words leading to false memories also tended to be slightly more concrete, on average, than those that did not. Presumably, people could generate a visual image more easily for the more concrete words.

"At any rate, the remarkable finding is that brain activity during the study phase could predict which objects would subsequently be falsely remembered as having been seen as a photograph," Paller said.

The flip side is that memory for viewed photographs was often correct. People gave many correct responses for objects they indeed viewed. Brain activity produced in response to viewed pictures and measured with functional MRI also predicted which pictures would be subsequently remembered. Two brain regions in particular -- the left hippocampus and the left prefrontal cortex -- were activated more strongly for pictures that were later remembered than for pictures that were forgotten. These two brain areas have previously been understood to play a central role in memory.

The new findings directly showed that different brain areas are critical for accurate memories for visual objects than for false remembering -- for forming a memory for an imagined object that is later remembered as a perceived object. The neuroanatomical evidence furthermore sheds light on the mental mechanisms responsible for forming accurate memories versus false memories.

"In the case of the false remembering emphasized here, the false memories were created when vivid visual imagery was engaged and a mental image was produced," Paller said. "These mental images left a trace in the brain that was later mistaken for the trace that would have been produced had that object actually been seen."

Listed as on the study, the co-investigators are Brian Gonsalves, post-doctoral fellow, Stanford University, and Northwestern researchers Paul J. Reber, associate professor of psychology, Darren R. Gitelman, associate professor of neurology, Todd B. Parrish, associate professor or radiology, M. Marsel Mesulam, Ruth and Evelyn Dunbar Professor, and Kenneth A. Paller, professor of psychology. The Northwestern researchers are affiliated with the department of psychology, the Institute for Neuroscience, the Cognitive Neurology and Alzheimer's Disease Center, the department of neurology, the department of radiology and the Feinberg School of Medicine.

###


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.