Hopkins Verbal Learning Test - revised process variables as embedded measures of performance validity

This study was designed to examine the potential of process variables within the Hopkins Verbal Learning Test - Revised (HVLT-R) to serve as embedded performance validity tests (PVTs).

Archival data were collected from 52 adults referred for neuropsychological testing. Performance validity was psychometrically operationalized using a combination of free-standing and embedded PVTs.

Seven HVLT-R process variables individually correctly classified on average 77-79% of the sample. The process variables were combined into a single validity composite (PRO-7), which produced statistically and clinically superior signal detection performance, correctly classifying 83-85% of the sample. The PRO-7 was unrelated to performance on tests of cognitive ability but had extremely strong linear relationship with composite PVTs. Male sex and lower levels of education was associated with failing the PRO-7.

HVLT-R process variables were statistically and clinically significant predictors of psychometrically operationalized invalid performance and were associated with implausibly low (non-credible) scores on a range of neuropsychological tests. The PRO-7 is a promising alternative approach to traditional PVTs based on the method of threshold. Generalizability of the findings is constrained by the small sample size. Replications are needed to further examine the utility of HVLT-R process variables as PVTs.