As part of an effort to develop a primate model of human age-related memory dysfunction, performance by six rhesus monkeys 26 to 27 years of age was compared to that of six young adult monkeys (four to five years of age) on a trial unique delayed nonmatching to sample (DNMS) task. This task assesses the monkey's ability to identify a novel from a familiar stimulus over a delay and resembles closely clinical tests that are used to assess memory function in geriatric patients. The task was presented in three stages: acquisition, delays and lists. As a group, aged monkeys were impaired relative to the young adult group on all three conditions. However, within the aged group, individual cases of efficient performance were observed. Error analyses of item positions of the lists condition revealed the absence of enhanced performance for items presented at the end of a list by aged animals, suggesting an abnormal sensitivity to proactive interference. The finding of a recognition impairment with age is in parallel with studies of normal human aging and lends support to the notion that the rhesus monkey is a suitable animal model of human aging.