This study examined the association between maternal cocaine use and maternal behavior and tested a conceptual model predicting maternal insensitivity during mother-infant interactions. Participants included 130 mother-infant dyads (68 cocaine-exposed and 62 non-cocaine-exposed) who were recruited after birth and assessed at 4-8 weeks of infant age. Results of model testing indicated that when the effects of prenatal cocaine use were examined in the context of polydrug use, maternal psychopathology, maternal childhood history, and infant birth weight, only postnatal cocaine use and maternal depression/anxiety were unique predictors of maternal insensitivity during mother-infant interactions.