Abstract
This study was designed to evaluate the potential of the Boston Naming Test (BNT) as a performance validity test (PVT). The classification accuracy of the BNT was examined against several criterion PVTs in a mixed clinical sample of 214 adult outpatients physician referred for neuropsychological assessment. Mean age was 46.7 (SD = 12.5); mean education was 13.5 (SD = 2.5). All participants were native speakers of English. A BNT raw score ≤ 50 produced high specificity (.87–.95), but low and variable sensitivity (.15–.41). Similarly, a T score ≤ 37 was specific (.87–.95), but not very sensitive (.15–.35) to psychometrically defined non-credible responding. Ipsative analyses (i.e., case-by-case review of individual PVT profiles) suggest that failing these cutoffs was associated with zero false positives when all available PVTs were taken into account. Results are consistent with previous reports that the validity cutoffs on the BNT have high positive predictive power, but low negative predictive power. As such, they are useful in ruling in invalid performance, but they cannot be used to rule it out.