Abstract
Background Neuropsychological tests are influenced by culture, language, level of education, and literacy, but there are few cognitive tests of which the applicability in ethnic minority populations has been studied. Objectives The aim of this study was to assess the reliability and validity of the Visual Association Test (VAT), a test of visual association memory, in a non-Western, low-educated memory clinic population. Additionally, a modified version of the VAT using colored photographs instead of line drawings was studied (mVAT). Method Both the original VAT and the mVAT were administered to non-Western immigrants (n = 73) from 2 multicultural memory clinics in Rotterdam, The Netherlands, and a control sample of non-demented Turkish elderly (n = 14) with low education levels (32 and 29% illiterate, respectively). Results Both the VAT and the mVAT were able to discriminate persons with and without dementia (area under the curve: VAT, 0.77–0.88; mVAT, 0.85–0.95). The mVAT had more homogeneous item difficulty levels than the VAT. Administration of parallel versions of the VAT and the mVAT within the same person revealed higher scores on the mVAT (Z = −3.35, p = 0.001). Conclusions The mVAT is a reliable and valid measure of memory in non-Western immigrants. Clinicians and researchers should be aware that the memory performance of immigrants may be systematically underestimated when using tests with black-and-white line drawings, such as the original VAT.