The Collaborative Initiative on Fetal Alcohol Spectrum Disorders (CIFASD) is a multi-site, interdisciplinary project aimed at characterizing structural and functional deficits in FASD to enhance understanding of both dysmorphology and the neurobehavioral phenotype (Mattson et al., 2010a). A previous study from the CIFASD program used latent profile analysis (LPA) to classify subjects with AE and controls. Using 22 neuropsychological variables with medium to large effect sizes, a 2-class model best fit the data with an overall success rate of 92% accuracy in distinguishing FAS from non-exposed controls; 87.8% of FAS cases and 95.7% of controls were correctly classified. Similar results were found when the profile was tested using AE subjects without FAS and controls: overall classification accuracy was 84.7%; 68.4% of alcohol-exposed and 95% of controls were correctly classified. In these analyses, executive functioning and spatial processing measures were most sensitive to prenatal alcohol exposure (Mattson et al., 2010b). The specificity of the profile was not tested. The current study aimed to build upon this preliminary neurobehavioral profile to (1) improve our classification accuracy, particularly for alcohol-exposed subjects without FAS
