all of the associations, facts, and lists of facts stored in memory compete with
the target information for the attention of the search process (Anderson, 1976,
1983; McGeoch, 1942; Postman, Stark, & Fraser, 1968).
The ease with which a record-keeping theory explains interference in mem-
ory experiments is one of the most compelling sources of evidence for it. But
the explanation leads to a paradox (Smith, Adams, & Schoor, 1978). As we go
through life, the number of associations with elements in our memory should
continually increase. It follows, then, that over time we should become increas-
ingly inefficient at finding information stored in our memory. Becoming an ex-
pert would be especially difficult, because the expert learns many facts about
a set of concepts. Experts, then, would be expected to have an ever-increasing
difficulty in remembering information in their area of expertise. Obviously this
does not happen. The record-keeping theory’s explanation for interference
observed in many memorization experiments cannot easily explain the obvious
facts that an adult’s memory skill remains stable over time and that experts get
better, and not worse, at remembering information in their area of expertise.
The difficulty that the record-keeping theory of memory has with explaining
everyday observations about memory reminds us that explanations must have
ecological validity. That is, theories of memory should explain how memory
works in the actual environment in which we use our memory. Some memory
theorists, notably Ulric Neisser (Neisser, 1978), have argued that a lot of mem-
ory research does not look at memory in realistic settings and so may not be
ecologically valid. Neisser has urged the cognitive psychological community to
make more use of experimental paradigms that resemble real-life situations. I
have tried to include a fair number of such experiments in this chapter. Some
memory theorists, however, have complained that experiments that have
resembled real-life situations have not really uncovered any new principle of
memory (Banaji & Crowder, 1989; see articles in the January 1991 issue of the
American Psychologist). Perhaps, though, the contrast between the results of list
memorization experiments and the everyday observations that memory is sta-
ble over time and that experts have good memory constitutes a compelling ex-
ample of the importance of conducting ecologically valid research.
The constructionist theory is able to explain both the decline in memory per-
formance exhibited in the memorization experiments and the lack of decline in
memory observed in ordinary day-to-day situations or in experts. Memoriza-
tion of related lists of words should generally present difficulties because the
same elements would be used repeatedly to understand each new list. A subject
who must memorize two lists of state-city associations, for example, would
find that the connections among the cognitive elements used to understand and
recollect the first list would be reconfigured when the second list was studied,
thereby undermining memory for the details of the first list. Memorization,
therefore, should be poorer or slower if a person has to memorize several re-
lated lists than if a person has to memorize unrelated lists. Furthermore, the
repeated use of similar lists would make accurate and detailed reconstruction
of any one list difficult.
The stability in an adult’s memory skill occurs because the elements used to
understand experiences do not expand in number as a result of having many
experiences. Only the connections among elements change with experience.
Memory 347