Neuropsychological Assessment of Brain Damage: a Validation Study of the Mccarron-dial System Dissertation

The present study investigates the effect of brain damage on verbal-spatial-cognitive (VSC) and sensorimotor (SM) measures included in the McCarron-Dial System (MDS). The subjects include 141 brain damaged adults and 42 psychiatric controls. The following research questions are addressed: (a) Does the brain damaged group differ significantly from controls? (b) Are there significant differences among left, right, anterior, posterior, and diffuse brain damaged groups? (c) Do early onset, late onset, acute, and chronic damaged groups differ significantly? and (d) Does a cerebral palsy group differ significantly from a non-CP brain damaged group? Analyses of variance reveal significant differences between the brain damaged and control groups on 44 of 53 VSC and SM measures; 97% of the brain damaged group and 71% of the controls are accurately classified by MDS and SM measures using discriminant analysis. Significant differences between the left and right brain damaged group are observed on 27 of 44 MDS measures; however, only 4 SM left hemisphere indicators significantly differentiate among left, right, diffuse, and control groups; 60% of the subjects in these groups are accurately classified. Only 3 of 26 MDS measures differ significantly among the anterior, posterior, and control groups; 73% of these subjects are accurately classified by SM measures. Significant differences between the early and late onset groups are observed on 18 of 24 MDS VSC and SM measures, but no MDS variables differ significantly between the acute and chronic groups. All 53 MDS VSC and SM measures differ significantly between the CP and brain damaged (without CP) groups; the motor scores are relatively lower in magnitude than other factor scores. The results support the validity of the MDS for neuro-behavioral diagnosis, but supplementary information is needed to increase prediction accuracy. A key approach to interpretation may have advantages over statistical models. Further validation research is needed to extend the system's utility for neuropsychological diagnosis.