Kelley, C. M., Sahakyan, L. (2003). Memory, monitoring, and
control n the attainment of memory accuracy. Journal of Memory and Language, 48, 704-721.
What underlies memory accuracy? These researchers suggest
that accuracy is based on the quality of our memory in conjunction with our
ability to adequately exact control over a metacognitive memory monitoring
process. These researchers appeal to Koriat and Goldsmith’s model of memory and
posit that memory is divisible into retrieval, monitoring and control. These
various component processes contain addition processes and characteristics, but
for the sake of this review consider them more generally. Retrieval is the
ability of an individual to access a memory trace for a past event. In some
circumstances individuals may engage a generate recognize strategy. When doing
so they will retrieve a plausible candidate even when they are not able to
consciously recollect its presentation. Monitoring is the process of assigning
a probability that a retrieved candidate is accurate. Control is exacted over
these processes by setting the criterion for the acceptance of an item as old.
The researchers are interested in applying this model to the differences
between the abilities of young and older adults to accurate remember the past.
Experiment 1
120 individuals (60 young adults, 60 older adults)
participated in this experiment. Of the study items one third were related
(e.g., morning – evening), one third were unrelated, and one third were
deceptive. The deceptive items were tricky in that the cue had a high associate
that was perceptually similar to the word that was presented at study (e.g.,
dollar and doctor). Participants took the memory test under moderate or high
accuracy incentive conditions.
Participants studied a list of 60 pairs of words. They then
completed a memory test that contained 60 study items and 15 filler items. On
the test participants were presented with the first word in a pair as a cue and
were forced to generate the other word that was presented, even if this
involved guessing. If they found a cue word to be unfamiliar they were
instructed to say “new”. After the generation of each word, participants were
instructed to rate the likelihood of the item being correct. Then participants
were asked to report if this candidate words was indeed presented. They could
say the word or say pass.
Results
See table 1 for a summary of the relevant results. Of importance
is the finding that incentive did not influence the ability of older adults to
improve their memory performance, if anything the incentive made them worse at
accurately performing the memory task. In addition, younger adults had
increased accuracy for the deceptive items than older adults. Also, it is
important to note that although older adults accuracy increased from forced to
free recall, this change was accompanied by a reduction in the quantity of
accurate items recalled. The monitoring process of younger adults was more
effective than that of older adults for both control and deceptive items. The
researchers hypothesized that the differences between the abilities of younger
and older adults to accurately monitor their memories was due to the quality of
the memory available to the monitoring process. To test this prediction the
researchers conducted experiment 2.
Experiment 2
Younger adults studied pairs of words under full or divided
attention. The procedures used in this experiment were identical to those used
in the first with the exception that there was no manipulation of incentive.
Results
Young adults who studied items under divided attention had
decreased accuracy for control and deceptive items. In addition the young
adults in the divided attention condition had decreased monitoring ability. The
participants in the divided attention condition performed much like the older
adults in comparison to the participants in the full attention condition. This
was a manipulation of encoding. Thus the differences in monitoring observed in
this experiment should be the result of memorial quality and not control. It
follows that the explanation that memory quality could result in the
differences observed in experiment 1 has some merit.
Experiment 3
Younger and older adults were compared on their ability to
remember pairs of words. Half of the younger adults studied the pairs of words
under divided attention and the remaining participants studied the pairs of
words under full attention. In this experiment participants only completed a
free response cued recall test. They were not forced to provide a cue. This was
done to control for the fact that the generation stage (i.e., forced response
test) used in the previous two stages could have influenced the strategies used
by participants to complete the memory test. As before young adults in the full
attention condition had better accuracy than the younger adults in the divided
attention condition. In addition, the young adults in the divided attention
condition had better accuracy than the older adults.
Discussion
These researchers conclude that the differences observed in
these experiments demonstrate older adults to be deficient in their ability to
recollect the past and instead rely on familiarity in cued recall settings.
This results in the differences between younger and older adults in respects to
memory accuracy. They also suggest that the reliance on familiarity,
plausibility and gist feeds into the monitoring process of older adults and they
fail to recalibrate their monitoring process to account for the lack of
recollection of contextual details when determining is a candidate word was
presented previously. Recollection in more precise than these other forms of
memory and as such older adults should adjust their monitoring process in
accordance.