Age-related working memory deficits in the allocentric place discrimination task: possible involvement in cholinergic dysfunction Takefumi Kikusui* ,1 , Toshiyuki Tonohiro, Tsugio Kaneko Neuroscience Research Laboratories, Sankyo Co., Ltd., 2-58 Hiromachi 1-chome, Shinagawa-ku, Tokyo 140, Japan Received 6 July 1999; received in revised form 9 September 1999; accepted 5 October 1999 Abstract It is well known that learning and memory ability declines with aging. Age-related long-term changes in learning and memory ability in rats were investigated with the place navigation task and the allocentric place discrimination task (APDT) in a water maze using the same animals for each task. In a working memory place navigation task, aged animals could learn the location of the platform as well as when they were young, although strategy shifts were observed. In contrast, accuracy in the APDT significantly declined from 90% to 65% with aging. This impairment was ameliorated by an acetylcholine esterase inhibitor physostigmine at 22–23 months old. No amelioration was, however, detected in the same animals tested when they further aged to 26 –27 months old. These results suggest that the APDT performance is sensitive to age-related memory deficits and that this may be due to the cholinergic dysfunction. © 2000 Elsevier Science Inc. All rights reserved. Keywords: Acetylcholine; Aging; Allocetric spatial cognition; Water maze; Working memory 1. Introduction In humans, it is well known that spatial learning and memory is impaired with aging [10,41,44,46]. The spatial learning and memory of aged rats are also impaired in various spatial task tests such as the Morris water maze [11,15–19]. In the Morris water maze, rats learn the location of a submerged platform in a circular pool using extramaze cues and navigate themselves onto the platform to escape from the water. This is a particularly useful tool for assess- ing age-related memory deficits because the motivating stimulus, escape from water, does not require food or water deprivation. Nutrient restriction in appetitive tasks, such as the radian arm maze or the T-maze, may endanger the health of aged rats. By avoiding food or water deprivation the animals can be kept in good physiological condition, espe- cially for long-term tests to evaluate aging effects in indi- vidual animals. For these reasons, the Morris water maze has been used as a model for nonverbal tests of spatial cognition in animals [29,31,38,39,51]. Two types of memory, reference and working memory, have been studied in experimental animals [37]. Reference memory is trial-independent: the information once learned is relevant to every trial. Previous reports focused on age- related reference memory deficits in water mazes [7,8,11– 13,15,16,25,39,40]. In contrast, working memory is trial- dependent: the information is relevant to a limited number of trials only. The latter type of memory has a major tem- poral component; thus, it can represent short-term memory. It has been reported that spatial cognitive dysfunction in aged humans is severe in both working memory and short- term memory [2,4 – 6,14,20]. Therefore, it is preferable to assess the decline of working memory ability during long- term aging by using the same animals, who have already learned a task when they were young, as a model for human senile dementia. The place navigation task (PNT) is dependent on two systems; the path integration system (i.e., how to reach the platform) and the allocentric orientation system (i.e., where the animal is and where the platform is) as previously described [33,35,36,43,49]. We previously reported that the allocentric place discrimination task (APDT), which is de- pendent on the allocentric orientation system only, is capa- * Corresponding author. Tel.: +81-3-5841-5475. E-mail address: akikus@mail.ecc.u-tokyo.ac.jp (T. Kikusui) 1 Current address: Laboratory of Veterinary Ethology, Department of Animal Resource Sciences, Graduate School of Agriculture and Life Sci- ence, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan. Neurobiology of Aging 20 (1999) 629 – 636 0197-4580/00/$ – see front matter © 2000 Elsevier Science Inc. All rights reserved. PII: S0197-4580(99)00096-2