Animal models are essential resources in basic research and drug discovery in the field of Alzheimer’s disease (AD). As the main clinical feature in AD is cognitive failure, the ultimate readout for any interventions or the ultimate goal in research should be measures of learning and memory. Although there is a wealth of genetic and biochemical studies on proposed AD pathogenic pathways, the aetiology of the illness remains unsolved. Therefore, assessment by cognitive assays should target relevant memory systems without assumptions about pathogenesis. The description of several tests that are available for assessing cognitive functioning in animal models can be found in literature. Among the behavioural test, the novel object recognition (NOR) task is a method to measure a specific form of recognition memory. It is based on the spontaneous behaviour of rodents and offers the advantage of not needing external motivation, reward or punishment. Therefore, the NOR test has been increasingly used as an experimental tool in assessing drug effects on memory and investigating the neural mechanisms underlying learning and memory. This review describes the basic procedure, modifications, practical considerations, and the requirements and caveats of this behavioural paradigm to be considered as appropriate for the study of AD. Altogether, NOR test could be considered as a very useful instrument that allows researchers to explore the cognitive status of rodents, and hence, for studying AD related pathological mechanisms or treatments.