The utility of the recognition memory test and the graded naming test in monitoring neurological patients

Objectives. Repeated neuropsychological assessments are often used to monitor cognitive changes in neurological patients. In this study we document: test-retest reliability (TRR), group changes in performance and measures of intra-subject variability for two commonly used cognitive tests. Design. A large sample of neurological patients with either acute or degenerative neurological damage, and relatively static cognitive profiles, were retrospectively selected if they had been administered twice either the Recognition Memory Test (RMT) and/or the Graded Naming Test (GNT) within a 2-year interval. Results. Overall, the TRR of the RMT was reasonable in both patient groups, although slightly better in the neurodegenerative population. The GNT had excellent reliability in both patient populations. For both tests, the intra-subject variability was generally large and greater than that observed in healthy adults. Conclusions. Despite its reasonable TRR, the RMT's very large intra-subject variability makes it unsuitable for monitoring the gradual changes in episodic memory performance that may be associated with both acute and degenerative neurological conditions. In contrast, the GNT is a reliable test of nominal skills and is sensitive to gradual changes in performance over time.