Neurobiology ofAging, Vol. 14, pp. 635-636, 1993
Printed in the U.S.A. All fightsreserved.
0197-4580/93 $6.00 + .00
Copyright© 1993PergamonPress Ltd.
The Longitudinal Assessment of Recognition
Memory in Aged Rhesus Monkeys
MARK B. MOSS
Department of Anatomy and Neurobiology, Boston University School of Medicine, Boston, MA 02118-2394
THESE behavioral data come from a multidisciplinary study on
the neurobiology of aging in rhesus monkeys as part of a program
project funded by the National Institute on Aging to Boston Uni-
versity in collaboration with the Yerkes Regional Primate Re-
search Center in Atlanta. Morphological data from some of these
behaviorally characterized monkeys will be presented by Dr. Jo-
hannes Tigges, Dr. Alan Peters, Dr. Thomas Kemper, and Dr.
Douglas Rosene. Our behavioral data complement those presented
by Drs. Rapp and Bachevalier on age-related memory loss in the
monkey and reinforce the observation that performance in aged
monkeys is characterized by marked individual differences. Be-
fore presenting our behavioral findings, I would like to briefly
address the issue of what constitutes an old monkey. Published life
span data from three cohorts of rhesus monkeys at the Yerkes
Primate Center (a total of 763 animals) indicate that 50% survive
to the age of 16 years, 25% survive to the age of 25 years, and less
than 10% survive beyond the age of 30 years (4). Review of life
span curves for humans shows that half of the population is still
alive by the age of 80, whereas for monkeys, this point on the
survival curve corresponds to the age of 16. Accordingly, mon-
keys 25-29 years of age, the range of the monkeys that were used
in these studies, likely represent the oldest of the old.
One of the objectives of our program has been to maintain, to
the extent possible, parallel studies of nonhuman primates and
humans. As an example of this approach, one the behavioral tasks
we have developed to assess visual recognition memory in the
monkey, and one I will discuss today, has been used extensively
to assess memory function in human clinical populations. The
delayed recognition span task (DRST) is a visual recognition
memory task that can be administered using several different stim-
ulus classes. For the spatial condition of the DRST in monkeys,
the animal is initially presented with an 18-well test board with one
well baited and covered by a disk. The animal is simply required
to displace the disk to obtain the reward. The screen is lowered,
and during the 10-s inter-trial interval, a new identical disk is
placed on the board over another baited well; the previous disk is
left in place and is now unbaited. The animal must displace the
new disk to obtain the reward and for the first trial is performing
in effect, a nonmatching to position task. If the animal succeeds in
displacing the new disk the next trial, another disk is placed on the
testing board and the animal is required to displace this additional
disk. This sequence continues until the animal commits its first
error, i.e., displacing a previously rewarded disk. The number of
correct responses before the first error is committed constitutes the
"recognition span."
The color condition of the DRST is administered in a similar
fashion to that for the spatial condition. This time the monkey is
required to displace a disk of a particular color; on the subsequent
trial, a new color is added to the board, but this time the previously
presented stimulus is moved to a different location on the board to
preclude the possibility of using spatial cues. Thus, this form of
the DRST serves as a nonspatial visual recognition memory task.
As in the spatial condition, the sequence continues until the animal
commits the first error and the number of correct responses until
the first error constitutes the recognition span. Ten such sequences
are given in a daily session, typically for 5 to 10 sessions.
For administration of this task to humans, the stimulus matrix
is expanded and verbal reinforcement is used instead of food re-
ward, but the test is otherwise carded out in a similar fashion to
that with monkeys. We have administered the DRST to normal,
young, adult, and aged individuals, as well as to a variety of
patients with age-related cognitive disorders such as Alzheimer's
disease and Parkinson's disease (1,3). We administered the spa-
tial and color DRST to a group of young (mean age 18.5 years)
normal adults and compared their performance to that of aged,
normal adults (mean age of 67.5 years). For the young adults,
the spatial span approached the ceiling of 16 positions, whereas
for the aged group, the mean spatial recognition span was only
11.5. A similar pattern emerged for the color condition. Hence,
the DRST and the recognition span it yields appears to be a useful
measure in detecting age-related changes in memory function in
humans.
In our earlier study of aged monkeys, we compared the per-
formance of four young adults (5-7 years) to normal aged (25-27
years) monkeys on both the spatial and color conditions of the
DRST. We found that aged monkeys were impaired relative to
young adults on both conditions of the task, a finding consistent
with the data from normal aged humans mentioned earlier. We
have since had the opportunity to assess the spatial and color
recognition spans in this same group of animals one year following
their initial assessment. Performance in the young adult group was
quite consistent. On the spatial span, three of the four animals
showed no change in performance and the fourth evidenced a
slight decline of less than 10%. For the color span, again, three
animals in the groups showed no change in performance over the
1-year interval. The fourth animal evidenced slightly improved
performance.
In the group of five aged monkeys, a very different pattern of
performance than that seen in the young adults was observed for
the two tasks. On the spatial condition, two animals showed no
change in performance, one showed a slight improvement, one
showed a mild loss and the last evidenced a marked loss. In con-
trast to their performance on the spatial condition, none of the
animals showed an appreciable loss in recognition span on the
color condition. In fact, two animals showed a slight improve-
ment. The apparent greater sensitivity to performance of the spa-
tial span as compared to color span by aged monkeys is similar to
that seen following ablations of the hippocampus (2). Whether or
635