Assessing object-recognition memory in rats: Pitfalls of the existent tasks and the advantages of a new test.

Studies of object-recognition memory in lab rats began in the late 1980s, using variants of the trial-unique delayed nonmatching-to-sample (DNMS) task. By the end of the 20th century, most investigators who wanted to study object-recognition in rodents had abandoned the DNMS task in favor of the novel-object-preference (NOP) test, mainly because the latter test is relatively easy to employ, whereas conventional DNMS tasks are not. Some concerns have been raised, however, about the internal validity of the NOP test as a method of measuring object-recognition abilities. We describe two experiments using a new DNMS procedure which requires considerably less training than the DNMS tasks of the 1980s and 1990s, and which cannot be subject to the same criticisms that have been leveled at the NOP test. In Experiment 1, rats were trained on the new modified-DNMS (mDNMS) task using short delays. Rats successfully learned the nonmatching rule in fewer than 25 trials, and they made accurate choices with retention intervals of up to 10 min. Experiment 2 examined a different group of rats' performance on the mDNMS task following long retention intervals (72 h, 3 weeks, and ~45 weeks). Rats made accurate choices on all retention intervals, even the longest retention interval of ~45 weeks. Overall, the findings demonstrate some benefits of an alternative approach to assess object-recognition memory in rats.