Utility of matrix reasoning as an embedded performance validity indicator in pediatric mild traumatic brain injury

Abstract

The present study examined the utility of the Wechsler Abbreviated Scale of Intelligence, Second Edition (WASI-II) Matrix Reasoning (MR) subtest in detecting suspect effort among a pediatric mild traumatic brain injury (mTBI) sample. The sample consisted of 526 clinically referred patients aged 8-16 years. Sixteen percent of participants failed the Medical Symptom Validity Test, which was used to categorize the sample into adequate and suspect effort groups. No background or injury-related variables differentiated the two groups. Comparisons between the adequate and suspect effort groups revealed significant differences in MR performance (p < .001; d = 1.46). A subsample (n = 206) where first item missed data were available was also analyzed; suspect effort participants missed their first item significantly earlier than adequate effort participants. In the entire sample, a MR cutoff T-score of 43 resulted in sensitivity of 57% and specificity of 90%. Receiver operating characteristic area under the curve for MR T-score was .804. Within this relatively high-functioning pediatric population with mTBI, MR appeared to have good utility as an embedded validity indicator, with classification statistics comparable to many of the most commonly used embedded indices used with adults.