Research Helps Understand Neural Machinery Of True And False Memories

In an article in the April 10, Proceedings of the National Academy of Sciences, Duke University neuroscientist Roberto Cabeza and his colleagues report experiments in which they asked subjects - while their brains were being scanned by MRI - to watch a videotape in which two speakers alternately recited lists of associated words, such as "water," "ice," "wet," "dark" and "freeze." The subjects were later presented lists of words that included the originally presented word, but also such related words as "cold" that were not actually presented before.

In earlier such memory experiments, by PNAS paper co-author Daniel Schacter of Harvard University, subjects who only listened to words showed a strong tendency to later falsely recognize non-presented critical words. However, they tended to rate true items higher than false items in terms of sensory details.

Said Cabeza, "This finding introduced a conundrum in false memory research. How can human participants believe in their illusory recollections and at the same time be able to differentiate them from veridical [true] recollections in terms of sensory detail?"

To explore in more detail the activity in the brain's memory region -- the medial temporal lobe -- during such recall events, the researchers designed the new experiment so that subjects under MRI would experience a richer sensory memory-inducing experience. Unlike the previous experiments in which words were simply heard by the subjects, the new experiments reported in PNAS featured an audiovisual presentation, in which each word was read on video by alternating male and female speakers.

The researchers then studied the activity of brain regions as the subjects were asked to perform a recognition test to tell whether a word was old or new. The list presented to the subjects included true, false and new words. As in the past, the subjects were able to reject new words, but falsely recognized false words related to the true words.

In analyzing brain activity during these tests, the scientists found a distinct difference in how areas of the medial temporal lobe responded to the true and false words. While the hippocampus responded to both true and false words, the parahippocampal gyrus was more activated for true than for false words.

According to the researchers, these findings indicate that the hippocampus was retrieving semantic information about the words, while the parahippocampal gyrus was retrieving perceptual information about the words, such as the speaker appearance and voice.

"These findings make particular sense, because we know that the parahippocampal gyrus receives inputs from various senses," Cabeza said. "And what's new about the study is that the differences we see are occurring within the medial temporal lobe, the most critical regions for memory processing. Until now, it has not been appreciated that different medial temporal regions treated processed true and false memories differently."

Cabeza and his colleagues plan further studies that will attempt to understand in more detail the newly appreciated heterogeneity of the memory region.

Besides Cabeza and Schacter, other co-authors of the paper are Stephen Rao and Andrew Mayer of the Medical College of Wisconsin, and Anthony Wagner of the Massachusetts Institute of Technology. Cabeza is a member of the Center for Cognitive Neuroscience at Duke.

The research was funded by the National Science and Engineering Council of Canada, the Alberta Heritage Foundation for Medical Research, National Institutes of Health, General Clinical Research Center of the Medical College of Wisconsin and the W. M. Keck Foundation.