The prevalence of pacifier use is high but when it occurs outside of the recommended sleep context, it becomes more controversial. Using 211 mother-infant dyads recorded as part of the Mercy Pregnancy and Emotional Wellbeing Study, we examined the maternal psychosocial predictors of pacifier use within an interaction task (i.e., ten minutes face-to-face followed by 30-minutes unstructured play). Predictors included maternal emotional availability measured with the Emotional Availability Scales; depression measured by the Structured Clinical Interview for the DSM-IV-TR Clinician Version; and maternal history of childhood trauma measured by the Childhood Trauma Questionnaire. An unadjusted odds ratio demonstrated that women classified as non-emotionally available to their infants were three-and-a-half-times more likely to use a pacifier. Multivariate logistic regression including all maternal psychosocial predictors demonstrated that even when adjusting for cessation of breastfeeding, maternal emotional availability remained the only significant predictor of pacifier use. This is the first time that predictors of pacifier use have been examined with a sample of clinically depressed women, as well as women with childhood trauma history. The results provide preliminary evidence that women who are not emotionally available might be more likely to rely on a pacifier during mother-infant interaction.