Evaluating the accuracy of the Wechsler Memory Scale-Fourth Edition (WMS-IV) logical memory embedded validity index for detecting invalid test performance. Issue 4 (4th July 2019)

• Record Type: Journal Article
• Authors: Soble, Jason R.; Bain, Kathleen M.; Bailey, K. Chase; Kirton, Joshua W.; Marceaux, Janice C.; Critchfield, Edan A.; McCoy, Karin J. M.; O'Rourke, Justin J. F.
• Is Part Of: Applied neuropsychology. Volume 26:Issue 4(2019)
• Page Start: 311
• Page End: 318
• Publication Date: 2019-07-04
• Subjects: Advanced Clinical Solutions (ACS) -- neuropsychological Assessment; performance Validity -- psychometrics -- Wechsler Memory Scale-Fourth Edition (WMS-IV); Neuropsychology -- Periodicals; Cognition disorders -- Periodicals; Brain -- Diseases -- Diagnosis -- Periodicals; Brain -- Imaging -- Periodicals; Brain -- Diseases -- Treatment -- Periodicals; Diagnostic Imaging -- Periodicals; Neuropsychological Tests -- Periodicals; Adult; 616.8
• DOI: 10.1080/23279095.2017.1418744
• Languages: English
• ISSNs: 2327-9095

Abstract

ABSTRACT: Embedded performance validity tests (PVTs) allow for continuous assessment of invalid performance throughout neuropsychological test batteries. This study evaluated the utility of the Wechsler Memory Scale-Fourth Edition (WMS-IV) Logical Memory (LM) Recognition score as an embedded PVT using the Advanced Clinical Solutions (ACS) for WAIS-IV/WMS-IV Effort System. This mixed clinical sample was comprised of 97 total participants, 71 of whom were classified as valid and 26 as invalid based on three well-validated, freestanding criterion PVTs. Overall, the LM embedded PVT demonstrated poor concordance with the criterion PVTs and unacceptable psychometric properties using ACS validity base rates (42% sensitivity/79% specificity). Moreover, 15–39% of participants obtained an invalid ACS base rate despite having a normatively-intact age-corrected LM Recognition total score. Receiving operating characteristic curve analysis revealed a Recognition total score cutoff of < 61% correct improved specificity (92%) while sensitivity remained weak (31%). Thus, results indicated the LM Recognition embedded PVT is not appropriate for use from an evidence-based perspective, and that clinicians may be faced with reconciling how a normatively intact cognitive performance on the Recognition subtest could simultaneously reflect invalid performance validity.