Stress can fool our memories, study finds

Originally published in The Arizona Daily Star, 2000 Tate Williams

Stress muddles memories, according to a UA study that raises doubts about eyewitness testimony in criminal cases.

Results of a standard word-memory experiment suggest that stress increases the likelihood that people will remember hearing words they actually did not hear. The participants in the study tended to remember the general themes of the words they had heard but confuse the details.

"Be really careful on not depending on the details," said Lynn Nadel, head of the University of Arizona psychology department and co-author of the study with graduate student Jessica Payne.

"Eyewitness testimony is open to significant error."

The study suggests that with stress, reconstructed details of the past become more fragmented and fictional, with false memory occurring nearly 80 percent of the time, compared with 60 percent in unstressed subjects.

The results fuel the idea that, while memory serves its purpose in our lives, it's less like a tape recorder and more like a rough reconstruction of fragmented pieces.

False memory has become a controversial topic in clinical and legal circles since adults in therapy apparently began recalling abuse in their childhoods. Also, new forensic techniques, such as DNA identification, have shown that eyewitness testimony isn't always reliable.

Nadel and Payne are quick to point out that their study does not support either side of the "false memory syndrome" debate.

While the suggestion that memory's details are sketchy is not directly applicable to real-world situations, it's not too large a jump to extend these results to other memories, Nadel said.

"There will be more research, and there needs to be more research. These things are somewhat artificial," Nadel said, referring to the laboratory situations Nadel considers an early step in stress research.

The experiment Payne conducted asked 84 subjects to listen to a series of words - candy, sour, sugar, bitter and cake, for example - and then take a test on the words. This test adds a "lure" - a related word that was not mentioned, like "sweet."

But before that, about half of the subjects, mostly college freshmen, were thrown into a situation that turns even a composed person into a sweaty-palmed mess.

Each "stressed" subject was told to prepare a short speech, which points out his or her best qualities.

The deadpanned experimenter then put the subjects on a stage before 1,000-watt lights, a camera and a one-way mirror that they were told hides "trained investigators" of the improvised speech. All along, the stressed subject was prodded by Payne over a microphone and loudspeaker.

The subject was then told to repeatedly subtract 13 from 1,022 aloud without stopping or making an error. This created stress that lingered in the brain during the memory test, she said.

Sixty percent of the unstressed people said the word "sweet" was on the list, even though it wasn't.

The stressed subjects recalled hearing "sweet" 77 percent of the time, but they did so without any more hesitation or uncertainty than when identifying words that were on the list.

"We need to appreciate the fact that memory is very fragile," said Larry Squire, a psychiatry and neuroscience professor at University of California at San Diego. "The paradigm itself suggests that you can create memory."

It's unclear whether the experiment reveals the workings of the brain in this matter, but it does demonstrate a symptom of stress, said Squire, who pioneered work on the hippocampus, a memory related part of the brain.

"It's useful because reports of eyewitness testimony can be so compelling," he said.