Detecting Response Bias on the MindStreams Battery

Empirically derived algorithm for performance validity assessment embedded in a widely used neuropsychological battery: Validation among TBI patients in litigation

“…The current study also aimed to cross-validate the findings of Hegedish et al (2012) among simulating participants. Hegedish et al (2012) compared coached simulator and nonsimulator (healthy and cognitively impaired) control groups and found performance patterns associated with high-NRB on three parameters of the staged information processing speed test.…”

“…Hegedish et al (2012) compared coached simulator and nonsimulator (healthy and cognitively impaired) control groups and found performance patterns associated with high-NRB on three parameters of the staged information processing speed test. In the present study we repeated the multivariate discriminant function analysis on the patients' data (DFA), but it did not yield a significant discriminant effect.…”

“…Logistic regression and receiver operating characteristic (ROC) were utilized in order to construct a classification model for the variables that best discriminate between low-NRB and high-NRB. Multivariate discriminant function analysis (DFA) was conducted in order to cross validate the results of Hegedish et al (2012), final model formula (information processing speed test: accuracy; response time, RT; and accuracy/response time × 100). All analyses were conducted with alpha level set at p < .05, using SPSS statistical software (SPSS 20, 2011;Chicago, IL).…”

“…Multivariate discriminant function analysis (DFA) was conducted in order to cross-validate the model developed in the previous study (Hegedish et al, 2012), which included three variables from the information processing speed test: accuracy, RT, and accuracy/response time × 100. The results of the DFA on the present cohort did not yield a significant discriminant effect, Wilks's lambda λ = .985, χ 2 (3) = 1.083, p > .05, between the low and high likelihood for NRB among the patients of the present study.…”

“…Given the necessity of evaluating negative response bias on a standardized computerized neuropsychological battery (NeuroTrax) and the advantage over stand-alone instruments for performance validity estimation, we devised the present study. In an earlier study (Hegedish, Doniger, & Schweiger, 2012), we attempted to create such an algorithm based on scores obtained from the NeuroTrax battery, but the participants were students instructed to simulate the performance of brain-damaged patients. Although that algorithm demonstrated reasonable specificity and sensitivity, it is necessary to validate such an algorithm and determine its clinical utility in the clinical population.…”