Attention and impulsivity of prenatally substance-exposed 6 year olds were assessed as part of a longitudinal study. Most of the women were light to moderate users of alcohol and marijuana who decreased their use after the first trimester of pregnancy. Tobacco was used by a majority of women and did not change during pregnancy. The women, recruited from a prenatal clinic, were of lower socio-economic status, and over half were African American. Attention and impulsivity were assessed using a Continuous Performance Task. Second and third trimester tobacco exposure and first trimester cocaine use predicted increased omission errors. Second trimester marijuana use predicted more commission errors and fewer omission errors. There were no significant effects of prenatal alcohol exposure. Lower Stanford-Binet Intelligence Scale composite scores, male gender, and an adult male in the household also predicted more errors of commission. Lower SBIS composite scores, younger child age, maternal work/ school status, and higher maternal hostility scores predicted more omission errors. These findings indicate that prenatal substance use has an effect on attentional processes.