Malingering Scales for the Continuous Recognition Memory Test and the Continuous Visual Memory Test

Continuous Visual Memory Test (CVMT) and Continuous Recognition Memory Test (CRM) scores of 24 litigants with definite malingered neurocognitive dysfunction (definite MND) were contrasted with those of 27 patients with moderate (n = 12) and severe (n = 15) traumatic brain injury (TBI). Analysis with chi-square identified 20 CVMT and 33 CRM items that discriminated the definite MND from TBI cases. Receiver operating characteristic (ROC) area under curve (AUC) was 0.918 for the CVMT scale, and 0.962 for the CRM scale, p = 0.184. Cutting scores were derived yielding sensitivity and specificity of 0.83 and 0.889 for the 20-item CVMT scale, and 0.875 and 0.93 for the 33-item CRM scale. On cross-validation contrasting performance of 17 litigants with evidence of probable MND and non-litigating psychiatric (n = 14) and neurologic (n = 13) patients, ROC AUC was 0.779 for the CVMT scale, and 0.847 for the CRM scale, p = 0.447. Sensitivity and specificity were 0.35 and 0.93 for the CVMT, and 0.529 and 0.96 for the CRM, using the cutting scores from the derivation sample. ROC AUC differences contrasting values for the definite MND/TBI comparison to the probable MND/neurologic and psychiatric comparison approached significance for the CRM SVT scale (p = 0.059), and for the CVMT SVT scale (p = 0.082). These data suggest that the malingering scales for the CVMT and CRM are more sensitive to definite rather than probable MND, and these scales are better at ruling in than ruling out malingered visual memory deficits.