Constructionist theories, on the other hand, imply that there is no all-purpose
memory system. Memory is instead a byproduct of changes to the various
components of cognition that underlie perception, language, emotions, and so
on. From the perspective of the constructionist approach, there are no storage
and retrieval mechanisms whose efficiency varies from person to person. In-
stead, people vary with respect to how much they know about various domains
of knowledge. According to constructionist theory, the main reason some peo-
plehavebettermemoriesthanothersisthatsomepeoplehavemoreexpertise
in the domain of knowledge sampled by the test of memory. For example, a
baseball expert can use the knowledge that runners on second base often score
after a single to reconstruct that the home team scored a run in the previous
inning. However, baseball knowledge would not help the baseball expert re-
member, say, a passage about climate in South America.
The constructionist theory claims, then, that the best predictor of how well a
person remembers new information in some domain, such as baseball, is how
much knowledge the person already possesses about that domain. General in-
tellectual skills, especially skill at memorizing lists of information unrelated to
the domain, should not predict individual differences in memory for informa-
tion within some domain. If, instead, memory is an all-purpose system, it
would follow that performance on tests of memory and on general intellectual
skills would readily predict memory for new information.
The research supports the constructionist theory’s explanation of individual
differences in memory. Good memory for information within some domain is
primarily a function of expertise in that domain and not a function of any gen-
eral intellectual skill. Schneider, Korkel, and Weinert (1987) and Walker (1987)
found that subjects who scored low on a test of general aptitude but happened
to know a lot about baseball recalled more facts about a fictitious baseball game
than did subjects who scored high on the general aptitude test but knew very
little about baseball, and recalled as many facts as did high-aptitude subjects
who knew a lot about baseball. Kuhara-Kojima and Hatano (1991) found that
knowledge about music, but not performance on a test of memory for unrelated
words, predicted how many new facts subjects recalled from a passage about
music.
Merely possessing domain knowledge does not guarantee better memory for
new information in that domain, however. DeMarie-Dreblow (1991) taught
people about birds but found that the newly acquired bird knowledge did not
help subjects recall a list of bird names any better than subjects not given the
knowledge about birds. The knowledge has to be well-learned, and people
need practice using the knowledge in the context of reconstructing a memory
for the new information (Pressley & Van Meter, 1994).
For instance, Pressley and Brewster (1990) taught their Canadian subjects
new facts about Canadian provinces. Some subjects were given prior knowl-
edge in the form of pictures of some prominent setting in the province. By it-
self, this prior knowledge did not help subjects remember the new facts all that
much better than the subjects not given the prior knowledge. Other subjects
were given imagery instructions for which the subjects were to imagine the fact
occurring in a setting unique to the province referred to by the new fact. Imag-
ery instructions also did not help subjects all that much. However, subjects
Memory 331