Natural Settings, Memory in
NATURAL SETTINGS, MEMORY IN
The study of memory as it is used in natural settings is now an accepted part of the scientific study of memory, engaging the attention of many researchers. This is a relatively new development; the classical study of memory took little interest in naturalistic studies. In the century between 1880 and 1980, an emphasis on experimental control led most experimenters to use specially prepared tasks and materials in their work, even if—as often happened—those tasks seemed meaningless to their subjects. The study of memory in natural settings was largely ignored.
Two important exceptions to this trend are worth mentioning: eyewitness testimony and memory for stories. In the early 1900s, William Stern established a journal (Beitrage zur Psychologie der Aussage) devoted entirely to the study of testimony. In his experiments, Stern staged unexpected or dramatic events in the presence of groups of people who were then interrogated as if they had witnessed a real crime. Such testimony is surprisingly unreliable. Witnesses often give highly confabulated accounts of the event itself and are rarely able to describe the "criminals." Confidence is no guarantee of accuracy in such reports. Modern research confirms that eyewitnesses may be entirely wrong even when they are quite sure they are right. One documented way to improve eyewitness testimony is Edward Geiselman's "cognitive interview," in which the witness is asked to recall events in different orders and from different perspectives.
F. C. Bartlett's 1932 book Remembering included the first systematic research on memory for stories, a field that was then largely neglected until the 1970s. Since that time, however, Bartlett's concept of "schema" has been widely adopted not only in the study of story recall but wherever memory is facilitated by preexisting mental structures.
A related subject is the study of memory as it appears in oral poetry and oral tradition—for example, in the preliterate heroic epics of ancient Greece. For decades this field was the province of humanistic scholars and classicists, but David Rubin's 1995 analysis based on the principles of cognitive psychology has been very successful.
Naturalistic versus Laboratory Methods
A 1978 conference at Cardiff, Wales, brought together many researchers with an interest in practical aspects of memory. In the opening address Ulric Neisser attacked the traditional study of memory as largely unproductive and called for a new emphasis on more naturalistic studies. Since that time many such studies have been conducted, and several related conferences have been held. This work is not without its critics. Some proponents of more traditional memory research believe that naturalistic studies have little value and that scientific progress can result only from controlled laboratory experiments. In an article titled "The Bankruptcy of Everyday Memory" Mahzarin Banaji and Robert Crowder asserted that a decade of naturalistic research had produced no important findings. A symposium incorporating nine replies to Banaji and Crowder appeared in the January 1991 American Psychologist. Since that time the controversy seems to have abated, with research on memory in natural settings now recognized as a significant component of the study of memory as a whole.
Because laboratory experiments are necessarily limited in time, studies of very long-term memory almost always take advantage of natural settings. The most systematic studies of this kind have been conducted by Harry Bahrick and his associates, who have found that some forms of memory are very persistent. The ability to recognize yearbook photographs of one's high school classmates, for example, remains strong even thirty-five years after graduation. In contrast, recall of classmates' names declines steadily over the same period. Recall for the material actually learned in school (e.g., a foreign language) tends to drop off for the first five years, but after that there is little more forgetting over the next twenty-five. Interestingly, even very brief reminders may bring much of the apparently forgotten material to mind again.
Recall of Life Experiences
A number of psychologists have studied their own memories by making brief notes of one or more experiences each day for a prolonged period. Then, after a delay of months or years, they test their own recall of the recorded events. These "diary studies" have produced several clear results. For one thing, forgetting is not all-or-none: an event that seemed at first to be completely forgotten may be retrieved if more cues are provided. Another fairly obvious finding is that experiences rated as important are recalled somewhat better than unimportant ones. But importance turns out to be very difficult to rate: the true significance of an experience may depend on later developments and hence may not be apparent when it occurs.
Some scholars have suggested that memory is egotistically biased, that we recall our own actions as more praiseworthy (and perhaps our own experiences as more pleasant) than they really were. Such distortions certainly do happen and have been verified in research settings. Nevertheless, the diary studies confirm this tendency only in part. On the average, experiences rated as pleasant are indeed better remembered than unpleasant ones. But at least one study found that personally relevant unpleasant experiences were remembered best of all. There are probably individual differences in these types of bias.
In a different research tradition, Michael Ross has proposed a specific theory of the bias that appears when people try to remember their own past characteristics. (How severe were my headaches last month? How fast did I read before I took this study-skills course?) Such estimates are not so much recalled as deduced from the trait's current level (my headache today, my reading speed now) together with an implicit theory of its stability or tendency to change. When that implicit theory suggests that a trait is stable (my headaches are chronic, I always read this fast), its past value is "remembered" simply by reference to its present level. The situation is different when the implicit theory suggests that a change should have occurred. Thus people who have just been through a headache treatment program (or taken a study-skills course) will tend to "remember" their earlier pain levels as worse (or their earlier reading speed as slower) than they really were. After all, the treatment (or the course) must have done some good! This is only one of several possible sources of error in recalling medical information. Information that confirms patients' previous beliefs seems better remembered than information that gainsays those beliefs.
Memories Triggered by Cues
Another way to study memory for life events—pioneered by Francis Galton more than a century ago—is to provide random words as cues and ask subjects to report whatever personal experiences (if any) the cue brings to mind. When the number of memories retrieved is plotted on a log-log scale as a function of the time that has passed since the original event, an approximately linear forgetting function appears. There are, however, two significant departures from linearity in forgetting. One is that middle-aged and elderly individuals recall disproportionately many events from their years of adolescence and early adulthood (roughly from ages ten to twenty-five). This pattern is called the "reminiscence bump." The bump appears not only in memories of life experiences but also in the recall of music, books, and news events: apparently it is the most memorable period of life in every sense.
The other departure from linearity appears at the beginning of the forgetting function: few if any events are recalled from the first few years of life. The near absence of memories from early childhood, which Sigmund Freud called "infantile amnesia," has been the object of considerable interest in its own right. Current interpretations of this phenomenon have been much influenced by modern studies of young children's memory. Although children as young as two years have some ability to recall past experiences, memories at that age have little internal organization and require much cueing by adults. Perhaps it is because young children lack the narrative skills that are necessary to link individual memories to one another and to a develop life story that adults can recall so little from that period of life.
Some experiences are remembered so vividly and confidently that they seem unforgettable, almost as though the brain took a snapshot of the scene at that moment; such recollections are often called "flash-bulb memories." Most "flashbulb memories" are of unique and personal experiences, but some are public (in a sense): they capture the moment when one first learned of some surprising and significant public event. Psychologists have studied the "flashbulb memories" created by hearing the news of the assassinations of President Lincoln, President Kennedy, and Martin Luther King Jr. (among others) as well as of other dramatic events such as the explosion of the space shuttle Challenger, the verdict in the O. J. Simpson trial, and so on. They have found that despite the strong subjective conviction that typically accompanies such memories, they are not neessarily accurate.
False memories can also occur in other contexts, often as a result of suggestion. Such suggestion can take many forms, but one particularly important context is psychotherapy. It is now well established that false memories of childhood sexual abuse can be produced by suggestion during therapy and that this can have serious consequences. To be sure, not all recollections of childhood abuse are false; many are all too valid. Unfortunately, valid and invalid memories are not distinguishable by any simple internal characteristic. The only way to establish the accuracy of a given recollection is with external evidence, which is not always available.
Bahrick, H. P. (1984). Semantic memory content in permastore: Fifty years of memory for Spanish learned in school. Journal of Experimental Psychology: General 117, 1-29.
Banaji, M. R., and Crowder, R. G. (1989). The bankruptcy of everyday memory. American Psychologist 44, 1,185-1,193.
Bartlett, F. C. (1932). Remembering. Cambridge, UK: Cambridge University Press.
Geiselman, R. E., Fisher, R. P., MacKinnon, D. P., and Holland, H. L. (1985). Eyewitness memory enhancement in the police in terview: Cognitive retrieval mnemonics versus hypnosis. Journal of Applied Psychology 70, 401-412.
Gruneberg, M. M., Morris, P. E., and Syjes, R. N., eds. (1978). Practical Aspects of Memory. Chichester, UK: Wiley.
Ross, M. (1989). Relation of implicit theories to the construction of personal histories. Psychological Review 96 341-357.
Rubin, D. C., Rahhal, T. A., and Poon, L. W. (1998). Things learned early in adulthood are remembered best. Memory & Cognition 26, 3-19.