Reliability and convergent validity of the Rey-auditory verbal learning test (RAVLT) against working memory tasks in a Serbian community sample

The Rey Auditory-Verbal Learning Test (RAVLT; Rey, 1964) is a widely used neuropsychological instrument that has not been validated in Serbia. Our aim was to examine its internal consistency, and construct validity, including convergent and discriminant validity against other cognitive variables, as well as its associations with age and education, in a cross-sectional, correlational study. Two hundred subjects from the general population in Serbia (53% women; \mathrm{M_{age}} = 43.45 years, \mathrm{SD_{age}} = 16.32) completed a battery of tests via oral examination. In addition to RAVLT, participants completed three working memory tasks: Digit Span, Digit Ordering, and Letter Memory, as well as the Rey-Osterrieth Complex Figure Test (ROCF). Participants also provided information on gender, age, education, parents’ education, household income, and any strategies used on the RAVLT. RAVLT showed good internal consistency. Its scores were negatively correlated with age (r = -.44, p < .001) and positively correlated with participants’ and their parents’ education (r = .31, p < .001). Gender differences in RAVLT performance were moderate to large, with women outperforming men. Correlations with working memory tasks were lower than expected (up to r = .33), and discriminant validity against ROCF was only established when controlling for age ( \mathrm{r_{partial}} < .15). Half of the participants reported using strategies (more frequently women), which were associated with better RAVLT performance. The RAVLT is a reliable and valid instrument, although the constructs it measures may be more complex than originally assumed. Our study also highlights the importance of strategy use in memory tests.