Jueves 12 de Junio de 2008, Ip nº 233

Total recall … Or at least the gist
Por Tom Jacobs

Our memory is like an ear of corn. At least, that's what Valerie Reyna was taught in graduate school.

Its Forrest Gumpish feel notwithstanding, the metaphor seemed scientifically sound. After all, researchers had already concluded there are two distinct types of memory: Verbatim, which allows us to recall what specifically happened at any given moment, and gist, which enables us to put the event in context and give it meaning.

"We were taught you extracted the gist from the verbatim memory," recalled Reyna, an experimental psychologist and former senior research adviser to the U.S. Department of Education. "It was like husking an ear of corn. You threw away the husk, which was the verbatim, and you kept the gist, which was the kernel of meaning."

There it was: Neat. Simple. Agrarian.

And also, as Reyna discovered over decades of subsequent research, wrong.

After conducting numerous studies with her partner, psychologist Charles Brainerd, Reyna concluded that verbatim and gist memory are separate, parallel systems. So separate, in fact, that "there is some evidence" they occupy different sections of the brain.

Reyna and Brainerd's hypothesis, which they call "fuzzy trace theory," explains how we can "remember" things that never really happened.

When an event occurs, verbatim memory records an accurate representation. But even as it is doing so, gist memory begins processing the information and determining how it fits into our existing storehouse of knowledge. Verbatim memories generally die away within a day or two, leaving only the gist memory, which records the event as we interpreted it.

Under certain circumstances, this can produce a phenomenon Reyna and her colleagues refer to as "phantom recollection." She calls this "a powerful form of false alarm" in which gist memory — designed to look for patterns and fill in perceived gaps —creates a vivid but illusory image in our mind.

Mental snapshots soon fade; what lingers are our impressions of an occurrence, which are shaped by the meanings we attach to it. If you doubt it, ask Sen. Hillary Clinton, who is still the butt of jokes for her insistence she came under sniper fire at a Bosnian airport.

"What she did (in misremembering the event) is very common, very ordinary," Reyna insisted. "People assume motivation, but you don't need it (to explain what she did). It happens to people every day. They think they're recalling something accurately, but they're not."

Reyna and Brainerd — both professors in the department of human development at Cornell University — summarized the research in this arena in their 2005 book The Science of False Memory, published as part of the Oxford Psychology Series. It spawned a series of follow-up studies, many of which are summarized in a paper they just published in the journal Psychology Bulletin. A number of others are ongoing.

"We're looking at a number of things, including the effect of emotion on memory — how emotion interacts with your interpretation of events," Reyna said. "Does arousal interfere with your encoding of memory? Does it ‘stamp it in,' as some of the neuroscience literature suggests? The effect might be more complex than that."

One question that can't be answered in the lab is why, in evolutionary terms, we would develop two separate memory systems. Reyna, who has given this considerable thought, noted that if all we had was our rapidly fading verbatim memory, "it would be very hard to function — especially in an oral culture. Cognition appears to be engineered around gist memory, which endures and is stable."

Consider the case of one of our prehistoric ancestors who is attacked by a saber-toothed tiger but manages to escape before being eaten. Verbatim memory would tell him precisely where the altercation took place, exactly what the tiger looked like and what tree he climbed to get beyond the animal's reach. Gist memory would tell him: "Tigers are dangerous. If I go walking in the forest after dark, I'd better bring my spear."

The first would be interesting; the second, essential. As Reyna wryly noted, "You don't have to count the stripes to know the tiger is bad."

So gist memory allows us to make snap decisions. But life does not always follow familiar patterns, and harm can result when we discard evidence that doesn't fit our assumptions. As the new paper notes, reports of overworked emergency room physicians misremembering patients' symptoms, and prescribing ineffective or even harmful "treatments," are not uncommon.

False memories arguably do the most damage in courtrooms, where jurors often consider eyewitness testimony particularly convincing evidence. The good news is there are clues that indicate whether a witness is remembering a situation accurately.
"There are things you can look for," Reyna said. "There are circumstances that lend themselves more to accurate testimony" —such as actually visiting the scene of the crime.

"To their immense credit, the courts are very open to the science," she added. "Judges have become more and more educated about this. We have a long way to go, but we're moving in (a positive) direction.

"I recently wrote a chapter of a handbook for lawyers. Several of my colleagues told me, ‘You'll have to dumb that down.' But the reaction was exactly the opposite. Lawyers wanted to read the original article and learn about the methodology. I was surprised and pleased."

One of Reyna and Brainerd's most controversial courtroom-related findings was released last year. It suggested the testimony of children might be more accurate than that of adults.

This belief was supported by an experiment in which they presented lists of words to groups of first-, fifth- and ninth-graders. Most but not all of the words were related by category; one list consisted of names of animals while another featured pieces of furniture.

After a short break, the kids were given new lists of words belonging to the same categories and asked to identify which were repeats. Consistently, the older children had more false memories, checking off words they thought they had seen earlier, but in fact had not.

Reyna was not surprised by these results. She noted that young children are forced to rely on verbatim memory, since their gist memory — that ability to connect a series of experiences into a meaningful pattern — is still in its formative stage.

"You turn a corner at age 11 or 12," she said. "A spontaneous connecting of the dots goes on. That seems to be a turning point."

Reyna cautioned, however, that many factors are involved when it comes to court testimony by youngsters. While their memories may indeed play fewer tricks on them, "Kids are much more susceptible to suggestibility from social factors," she said. In other words, they may feel compelled to say what a trusted adult wants or expects them to say.

"We hope that the ultimate effect (of this research) will be the same as that of DNA, which both implicates and exonerates people," Reyna said. "It just does it more accurately. I hope there will be a national movement to bring these practice guidelines on line."

To that end, "we just started a new program here at Cornell," Reyna said. Its goal is to create legally knowledgeable social scientists who "will be positioned to be leaders in making sure the law and public policy is informed by objective science. We want lawyers and judges to really understand this material, rather than relying on experts to tell them what it means."

Reyna called the new program — which is technically a graduate concentration — "Psychology, Law and Human Development." But a check of the university's Web site reveals the actual name is "Law, Psychology and Human Development." Inadvertently but elegantly, she reinforced her central point. Memory plays tricks even on those who study memory.


  08/05/2008. Miller-McCune Magazine.