Adaptation and initial validation of the Arabic version of the Word Memory Test (WMTARB)

Abstract

Background The feigning of cognitive impairment is common in neuropsychological assessments, especially in a medicolegal setting. The Word Memory Test (WMT) is a forced-choice recognition memory performance validity test (PVT) which is widely used to detect noncredible performance. Though translated to several languages, this was not done for one of the most common languages, Arabic. The aim of the current study was to evaluate the convergent validity of the Arabic adaptation of the WMT (WMTARB) among Israeli Arabic speakers. Methods We adapted the WMT to Arabic using the back-translation method and in accordance with relevant guidelines. We then randomly assigned healthy Arabic speaking adults (N = 63) to either a simulation or honest control condition. The participants then performed neuropsychological tests which included the WMTARB and the Test of Memory Malingering (TOMM), a well-validated nonverbal PVT. Results The WMTARB had high split-half reliability and its measures were significantly correlated with that of the TOMM (p < .001). High concordance was found in classification of participants using the WMTARB and TOMM (specificity = 94.29% and sensitivity = 100% using the conventional TOMM trial 2 cutoff as gold standard). As expected, simulators' accuracy on the WMTARB was significantly lower than that of honest controls. None of the demographic variables significantly correlated with WMTARB measures. Conclusion The WMTARB shows initial evidence of reliability and validity, emphasizing its potential use in the large population of Arabic speakers and universality in detecting noncredible performance. The findings, however, are preliminary and mandate validation in clinical settings.