Take a few minutes and study this list of words: bed, rest, awake, tired, dream, wake, snooze, blanket, doze, slumber, snore, nap.
You'll be asked about those words later. Meanwhile, here is some background on an important legal case called State of New Jersey v. Michaels. In 1988, Kelly Michaels, a preschool teacher at the Wee Care Day Nursery in Maplewood, N.J., was convicted on 115 counts of sexually abusing 19 children in her care. The 23-year-old Michaels was found guilty based entirely on the testimony of her alleged victims, who were 3-to-5-years-old. The children's memories were quite explicit and vivid and shocking: Michaels playing the piano in the nude, Michaels licking peanut butter off the children's genitals, and so forth. After a nine-month trial and 13 days of jury deliberation, Michaels was sentenced to 47 years in prison.
Michaels's guilty verdict was overturned on appeal six years later, based in large part on the testimony of psychologists about the suggestibility of young children. Long after Michaels went free, however, her high-visibility case remains an important legal landmark, legitimizing the wisdom that the testimony of very young children is and should be suspect. And the science has tended to bolster that legal position, namely, that memory always becomes more accurate with age, that adults are less susceptible to false memories and that kids are on balance untrustworthy witnesses.
But is this true? New science is questioning some of the received wisdom about false memories. Let's go back to that word list for a minute. Take out a piece of paper and, without looking back, write down as many words as you can recall from the list. That's what you would be asked to do if you were in a psychology laboratory, only you'd do it many times with different lists. If you are like most people, you would recall some words and forget others, but you might also recall a word that was never on that list: sleep. Why is that? Well, according to a new and powerful theory of human reasoning called "fuzzy trace theory," many of the memories we put down (and later recall) are not precise, literal memories at all. They are "fuzzy," merely the "gist" of meaning: that list may not include the word sleep technically, but it is all about sleep.
Two teams of psychologists at Cornell University, who were among the experts whose brief freed Kelly Michaels, have been exploring the nature of memory and meaning in both children and adults, to see how storage, sorting, recall and meaning-making change over time. The studies go directly to false memory and suggestibility, and have important implications for the frequent criminal trials that pit children's and adult's memories against each other, including trials for sexual abuse.
Both of the studies have to do with what's called semantic representation, which is just jargon for how we make sense of the thousands upon thousands of bits of information we encounter every day. In the first study, Charles Brainerd and Valerie Reyna used the vocabulary test mentioned before to compare children's and adults' propensity to create false memories. Coming up with the word "sleep" is a rudimentary kind of false memory—the brain creates it out of whole cloth to make sense of all those other sleep-related words. Brainerd and Reyna predicted that, contrary to doctrine about cognitive development, older children and adults would actually be more susceptible to false recall as their "gist" memories matured. And that is exactly what they found. The older the participants in the study, the more likely they were to "remember" words that they never saw.
Now, gist thinking is a good thing. It allows us to generalize and make connections. Indeed, we couldn't get through life without it, because otherwise we would have to remember every detail of our experience in order to make even the simplest decision. But Brainerd and Reyna's point is that, under some circumstances, such mature thinking can trip us up. Unhappily, those circumstances occur not infrequently in real life and can have life-altering consequences, as in cases of serial abuse, when the connections between one fact and another may or may not be true.
The other Cornell study reinforces this view, using a very different method for exploring childhood memory. In this experiment, Stephen Ceci and his colleagues had children of different ages form semantic maps among words. For example, the kids might see the words eagle, bear and robin, and they would have to decide the best way to group them for meaning. Ceci predicted that the psychological "distance" between the words would determine how suggestible the kids would be later on when he tried to trick them into remembering a similar but novel word.
This turned out to be true, but an even more important finding was that older participants were subject to different kinds of false memories than the younger ones. It turns out they have different categories in their heads. For example, where younger kids tend to group eagle and robin as birds, older kids typically group eagle and bear as predators. They have a predator dimension in their memories that young kids lack. As a result of this fundamental difference in memory organization, older and younger children (and presumably adults) make different kinds of mistakes with recall.
These are not just word games. These basic psychological experiments, both reported in recent issues of the journal Psychological Science, have everything to do with how we make meaning in the world. And how we sometimes distort it. It doesn't mean that very young children are always more reliable witnesses than teenagers or adults. They were clearly unreliable in the Wee Care case. But it does mean that they might be more accurate in their memories, depending on the nature of the events and evidence and testimony. An estimated 20,000 children testify in sexual-abuse trials every year, and juries are typically told that, given the choice, adult memory is more accurate than childhood memory. That instruction now requires some serious rethinking.