THE sensation of 鈥渞emembering鈥� something that you think you have
heard鈥攂ut didn鈥檛鈥攊s much the same as remembering something you
really did hear. But your brain knows the difference, according to researchers
in the US who have shown that different areas of the brain are active in the
recognition of real and false memories.
Daniel Schacter, a psychologist at Harvard University, and colleagues in
Arizona, Ohio and Missouri, asked 12 women to listen as 24 lists of 20 words
were read aloud to them. Ten minutes later, they presented the subjects with
written lists containing some words they had heard and other, related words they
had not. For instance, subjects who heard the words 鈥渟harp鈥�, 鈥減oint鈥�, 鈥減rick鈥�
and 鈥減ain鈥� on the original list found the word 鈥渘eedle鈥� on the written test
list. When the women were asked which words they had heard before, they were
almost as likely to identify words they had not heard as words they had.
While the women were deciding which words they had heard, Schacter and his
team used positron emission tomography (PET) to image the flow of blood to their
brains. In a paper to be published in next month鈥檚 issue of the journal
Neuron, the scientists report that the PET scans showed that both accurate
and false recognition used areas of the brain near the hippocampus that are
associated with memory. But real memories activated another part of the brain as
well: regions in the left superior temporal lobe that are responsible for
auditory information.
Advertisement
When subjects were evaluating words they had not heard on the original list,
they also showed greater activity in the frontal regions, which are believed to
be involved in the recovery and monitoring of memories, and in a region of the
cerebellum associated with detecting errors. 鈥淥ur hypothesis is that subjects
were engaged in an internal struggle about these words,鈥� says Schacter. 鈥淭hey
are doing extra monitoring or verifying.鈥�
A PET scan takes about a minute to complete, so the researchers had to
examine the women鈥檚 responses to blocks of heard and unheard words, not to
individual words. Nevertheless, the findings are still 鈥渧ery exciting鈥�,
according to Morris Moscovitch, a neuropsychologist at the University of Toronto
who specialises in memory.
Schacter is cautious about extrapolating this 鈥渂asic science finding鈥� to the
courtroom by using the PET scanner as a lie detector. The sensory details that
accompany a real memory could fade within a few days of the event, he says.
Alternatively, people who have convinced themselves that something has happened
by thinking about it over and over may come to encode invented sensory details.