Embedded Performance Validity Indicators Within the California Verbal Learning Test, Children’s Version

Abstract

To date, few studies have examined the use of embedded performance validity indicators in pediatric populations. The present study examined the utility of variables within the California Verbal Learning Test, Children’s Version (CVLT-C) in detecting noncredible effort among a pediatric mild traumatic brain injury sample. The sample consisted of 411 clinically referred patients aged 8–16 years. A total of 13% of the participants failed both the Medical Symptom Validity Test and at least one other performance validity measure. No demographic or injury-related variables differentiated the noncredible and adequate effort groups. The noncredible group performed significantly worse than the adequate effort group across a majority of CVLT-C variables. Logistic regression analysis revealed that the Recognition Discriminability (RD) score was the most robust in predicting noncredible effort. Among this relatively high-functioning sample, an RD cutoff z-score of –0.5 resulted in sensitivity of 55% and specificity of 91%. A more conservative RD cutoff z-score of –1.0 resulted in sensitivity of 41% and specificity of 97%. These findings are comparable to the classification statistics found for many embedded indicators in the adult literature. Although only moderately sensitive, the RD score on the CVLT-C appears to have good utility in identifying noncredible effort in a relatively high-functioning pediatric mTBI population.