By now, it should be evident that a great deal has been learned about the precise variables that control children's false memories and about how the mix of those variables leads to different patterns of developmental change in false memory. A key feature of the memory illusions that have been explored—narrative tasks in the constructivist tradition and word tasks in the verbal learning tradition—is that they are illusions of meaning: that false recall and false recognition are apt to occur when rememberers rely on meaning information that was instantiated by target materials. However, current theoretical analyses of this situation propose that more is involved. Those analyses are dominated by an opponent-processes principle, which is simply the idea that rememberers store and retrieve two basic forms of information, such as verbatim traces of the surface form of target materials and gist traces of materials' meaning content, with one type of processing supporting false memories ("Yes, I ate a hot dog, because that's what people usually eat at baseball games") and the other suppressing false memories ("No, I didn't eat a hot dog, because I clearly remember eating that cold, soggy slice of pizza") We have seen that the opponent-processes conception has been able to explain and, far more important, to predict a broad range of developmental findings. With narrative false memory, for example, we have seen that this conception was able to account for the conflicting age trends that have accumulated over the years in developmental studies of Bransford-Franks-type tasks. With false memory for words, we have seen that the same conception explains why Underwood-type tasks have usually produced age declines in false memory. The most important accomplishment, however, has been the prediction, some years before confirmation was obtained, of developmental increases in false memory for words. Such increases were predicted for tasks in which false-memory items access meanings that are repeatedly cued by target materials—most notably the DRM paradigm, but also tasks that involve repeated cuing of familiar taxonomic categories. Although this prediction is counterintuitive under the commonsense "memory gets better" view of development, it has been confirmed in so many recent studies that its validity is no longer in doubt. Crucially, the most recent studies in this literature have begun the process of drilling down to more precise hypotheses about the mechanisms that control age variability in false memory. In the end, it is findings from these tests of more precise hypotheses that will have the greatest yield in forensic work, as well as in theory development.
If some types of false memories can increase with age, this has important implications for a baseline assumption that the law currently makes about age variability in false memory. The classical finding, mentioned earlier in this chapter, that the incidence of false-memory declines between early childhood and young adulthood has figured prominently in many trials, but especially CSA cases, in which evidence was provided by child victims or witnesses (e.g., Ceci & Friedman, 2000). That finding has been used to challenge children's competence to testify in pretrial motions and pretrial hearings and to challenge children's courtroom testimony as being unreliable, all on the specific ground that research shows that their statements are far more likely than adults' to be tainted with false memories (Brainerd & Reyna, 2005). For example, a common occurrence in CSA cases is that there are two versions of events that bear on a defendant's guilt. One version is exculpatory and one is incriminating, and one version is provided by adults and the other by children. In such circumstances, the law requires juries to render a verdict by assessing the relative credibility of the contrasting stories. If research shows that children's memories are invariably more likely to be tainted with false information than adults', then, other factors being equal, the adult version should be assigned a higher credibility rating.
This is precisely the thrust of much expert testimony in CSA cases. In fact, such expert testimony has become so commonplace and the research that supports it has been so widely disseminated in the popular media that judges sometimes rule against allowing it to be presented in court, on the grounds that the heightened susceptibility of children to false memories is well understood by jurors and, therefore, expert testimony is not needed to educate them on it (McAuliff, Nicholson, & Ra-venshenas, 2007). However, a key implication of the more recent research that we reviewed earlier is that the law errs when it makes such a baseline assumption about the fallibility of children's memories. More recent research demonstrates that the direction of age variability in false memory depends on specific storage and retrieval factors. Therefore, in particular court cases, the scientifically appropriate course for expert witnesses is to analyze the circumstances of each case to determine which factors are present that are known to affect age variability in false memory and to educate juries (a) on the results of their analysis and (b) on the research literature that substantiates the factors that have been identified.
However, some readers may be disposed to be uneasy about the applicability of findings from research on laboratory tasks that involve narrative false memory to legal cases and, still more, about the applicability of findings from research on false memory for words. We can do no better than refer them to contemporary discussions of this question by Howe et al. (2004) and by Reyna et al. (2007), both of whom explore the question in connection with the more extreme of the two situations, false memory for word lists. This gist of their discussion runs as follows.
Superficially, to those who are steeped in the study of everyday memory it might seem that false memories for words have little or nothing to do with the real-life events that are the stuff of forensic work. Perception is not necessarily reality, however. It is often the case in science that differences that seem to be fundamental at first blush fade out upon deeper consideration, and indeed, tasks such as the DRM paradigm and the categorized-list paradigm share some of the most basic features of foren-sically relevant memories. For that reason, these paradigms may be quite relevant to such memories. In familiar real-life crimes, such as robberies or assaults, there are certain events that are characteristic of those crimes (e.g., fear, threatening statements, weapons), so much so that they are formulaic elements of crime novels and the scripts that writers produce for movies and television programs. Thus, when children are witnesses or victims of familiar crimes, they are exposed to collections of target events that are interconnected in meaning by virtue of the fact that, like hot dogs and home runs at baseball games, the events are prototypical of those situations. In the DRM and categorized-list paradigms, too (and in narratives, by the way), children are also exposed to multiple targets that are related in meaning. As we have seen, children falsely recall and falsely recognize unpresented exemplars of these shared meanings, these errors are affected by theoretically specified variables, and these errors display age trends that can be predicted on theoretical grounds. Considering the extensive data base on these paradigms, therefore, it is not surprising that children would have false memories of events that cohere with the gist of crimes, or that those false memories could vary with age.
Further, in CSA and other crimes of abuse and neglect, the phenomenon of encountering multiple exemplars of a meaning (the gist of the crime) can occur across multiple exposures to that crime (Powell, Roberts, Ceci, & Hembrooke, 1999; Reyna et al.,2007). A hallmark of such crimes is that they may occur in children's homes, with the perpetrators being children's caretakers or even other children. When they do, children may be victims or witnesses of multiple instances of a crime. Each instance of the crime exposes children to events that share meaning with prior instances but that are not exactly the same in key details as prior events. Moreover, each subsequent exposure is apt to cue memories for earlier events from prior exposures, particularly if children access stable gist memories and realize that the different experiences are connected. If children are aware that later events are not exact repetitions of earlier events but are related to earlier events, the fact that those different events were part of different episodes may become blurred, interfering with accurate verbatim memories of the details of the individual events.
This problem of interference with verbatim memory is exacerbated by the fact that when children have been victims or witnesses of multiple instances of index crimes, spaced over extended time intervals, investigative interviews and sworn testimony often will not occur until months or years after the initial instances. In such circumstances, their memory reports of what happened may be highly accurate (especially the gist of what happened) but their memory reports of when it happened will undoubtedly be highly inaccurate. Powell et al. have reported data showing that this extrapolation from word-memory studies is indeed borne out in experimentation on children's memory for everyday events that are of forensic significance, and other examples of forensically relevant data have been reviewed by Reyna et al. (2002).
Howe et al. (2004) raised the following important consideration: Because the types of paradigms that have been the foci of this chapter rely on target materials that have nothing to do with crimes or with physical or psychological trauma, it might be thought that children would be less likely to develop false memories for such crimes than for word lists or sentences. However, these crimes involve gists, strong gists in fact, as surely as the list "black, blue, brown,green,pink,purple, red, and white" does. Hence it is not clear what the experimentally verified, as opposed to speculative, basis for such an argument would be. Perhaps the most obvious basis that might be invoked is that the events of crimes are usually experienced in conjunction with high levels of anxiety and stress, which may inoculate memory against falsification. Indeed, some research has shown that stress and anxiety lead to greater memory accuracy (Christianson, 1992). As Reyna et al. (2007) pointed out, though, experimental findings on how anxiety and stress affect children's false memories are equivocal. Further, there is a good deal of evidence in the adult forensic memory literature that the anxiety and stress of crimes foment false memories, rather than suppress them (e.g., Kassin, 1997, 2001), and there is also a good deal of evidence that adults can develop false memories of complex life experiences, including crimes, that are fraught with anxiety and stress (for a review, see Brainerd & Reyna, 2005). Thus, the fact that crimes normally involve stress and anxiety apparently does not inoculate people against developing false memories.
Was this article helpful?
Parenting is a challenging task. As a single parent, how can you juggle work, parenting, and possibly college studies single handedly and still manage to be an ideal parent for your child? Read the 65-page eBook Single Parenting Becoming The Best Parent For Your Child to find out how. Loaded with tips, it can inspire, empower, and instruct you to successfully face the challenges of parenthood.