Using the Grooved Pegboard Test as an Embedded Validity Indicator in a Mixed Neuropsychiatric Sample with Varying Cognitive Impairment: Cross-Validation Problems

Abstract

Embedded validity indicators (EVIs) derived from motor tests have received less empirical attention than those derived from tests of other neuropsychological abilities, particularly memory. Preliminary evidence suggests that the Grooved Pegboard Test (GPB) may function as an EVI, but existing studies were largely conducted using simulators and population samples without cognitive impairment. In this study we aimed to evaluate the GPB’s classification accuracy as an EVI among a mixed clinical neuropsychiatric sample with and without cognitive impairment. This cross-sectional study comprised 223 patients clinically referred for neuropsychological testing. GPB raw and T-scores for both dominant and nondominant hands were examined as EVIs. A known-groups design, based on ≤1 failure on a battery of validated, independent criterion PVTs, showed that GPB performance differed significantly by validity group. Within the valid group, receiver operating characteristic curve analyses revealed that only the dominant hand raw score displayed acceptable classification accuracy for detecting invalid performance (area under curve [AUC] = .72), with an optimal cut-score of ≥106 seconds (33% sensitivity/88% specificity). All other scores had marginally lower classification accuracy (AUCs = .65–.68) for differentiating valid from invalid performers. Therefore, the GPB demonstrated limited utility as an EVI in a clinical sample containing patients with bona fide cognitive impairment.

Assessing the credibility of clinical presentations using performance and symptom validity tests: Current trends and future directions—Part II.

“…In contrast, perhaps because there are many more well-validated and widely used PVTs than SVTs Sweet et al, 2021), current PVT research seems to focus more on expanding the knowledge base on existing instruments, in the form of introducing new trials (C. Abeare et al, 2019;C. A. Abeare, An, et al, 2022;Kim et al, 2010;Lichtenstein et al, 2017;Rai et al, 2019), indicators or cutoffs (Arentsen et al, 2013;Ashendorf, 2019;Ashendorf et al, 2017Ashendorf et al, , 2021Blaskewitz et al, 2009;Bortnik et al, 2010;Boskovic et al, 2018;Boucher et al, 2023;Chang et al, 2023;Curtis et al, 2008Curtis et al, , 2009Deloria et al, 2021;Dunn et al, 2021;Erdodi et al, 2016Erdodi et al, , 2018Glassmire et al, 2019;Greve et al, 2009;Guise et al, 2014;M. J. Holcomb et al, 2022;Hurtubise et al, 2020;Johnson et al, 2012;C.…”

“…Abeare et al, 2019; C. A. Abeare, An, et al, 2022; Kim et al, 2010; Lichtenstein et al, 2017; Rai et al, 2019), indicators or cutoffs (Arentsen et al, 2013; Ashendorf, 2019; Ashendorf et al, 2017, 2021; Blaskewitz et al, 2009; Bortnik et al, 2010; Boskovic et al, 2018; Boucher et al, 2023; Chang et al, 2023; Curtis et al, 2008, 2009; Deloria et al, 2021; Dunn et al, 2021; Erdodi et al, 2016, 2017, 2018; Glassmire et al, 2019; Greve et al, 2009; Guise et al, 2014; M. J. Holcomb et al, 2022; Hurtubise et al, 2020; Johnson et al, 2012; C.…”