Fetal alcohol spectrum disorders (FASD) is a non-diagnostic term that encompasses the cognitive and behavioral deficits associated with prenatal alcohol exposure, including executive functioning, language, adaptive functioning, visuospatial processing, learning, and memory. At the severe end of the spectrum is fetal alcohol syndrome (FAS), which is defined by specific craniofacial features, growth deficiency, and central nervous system dysfunction. Not all children with FASD exhibit the criteria necessary for a diagnosis of FAS, but demonstrate comparable neurobehavioral delays. Both immediate and delayed memory deficits represent a significant impairment for individuals with prenatal alcohol exposure and can impact all aspects of life. Verbal, spatial, and facial memory deficits are the most commonly researched types of memory in this population, although the underlying mechanisms of these impairments remain unknown. Deficits may be inherent to general memory impairments or may be attributed to unique underlying mechanisms. This study investigated whether the following cognitive deficits contribute to domain-specific memory impairments in alcohol-exposed children: language functioning to verbal memory, visuospatial processing to spatial memory, and adaptive functioning to facial memory. Subjects were children (7-16 years) with prenatal alcohol exposure (AE, n = 63) and (CON, n = 53) with minimal or no alcohol exposure. Archival data was analyzed. Subjects were tested using the NEPSY-II to measure visuospatial processing and language functioning and the three memory domains: verbal, spatial, and facial memory. In addition, the Vineland Adaptive Behavior Scale-Second Edition (VABS-II) was administered to parents to assess the child's adaptive functioning. To determine group differences on verbal, spatial, and facial immediate and delayed memory, a 2 (group: AE and CON) _ 2 (condition: immediate and delayed condition) ANOVA was conducted separately for each memory domain. To evaluate the contribution of corresponding cognitive domains on its associated memory domain, separate hierarchical regression analyses were conducted. For each regression, group and covariates were entered in step 1, the related cognitive ability (mentioned above) was entered in step 2, and the interaction between cognitive ability and group was entered in step 3. Results showed that the AE subjects performed worse than the controls on verbal, spatial, and facial memory. As hypothesized language functioning predicted verbal memory, visuospatial processing predicted spatial memory and adaptive functioning predicted facial memory.