This study examines the cognitive profiles of infants born to HIV positive mothers in Zimbabwe. Caregivers with HIV exposed infants delivered in 30 clinics in two areas of Zimbabwe were recruited to the study. Of the 574 study participants, 562 caregiver-infant dyads with a biological HIV +ve mother and infant aged 0-24 months were interviewed. All infants were tested by a trained administrator for cognitive development on the Mullen Scales of Early Learning (MSEL). The Edinburgh Postnatal Depression Scale and Parental Stress Index-Short Form were completed by the mothers together with infant and caregiver socioeconomic characteristics. Linear regression models were used to relate cognitive development scores to maternal stress scores, maternal depression scores and infant HIV status adjusting for infant and caregiver characteristics, as well as socioeconomic factors. Higher maternal depression scores were associated with lower overall infant cognitive scores (adjusted mean difference (aMD) = -0.28; CI 95%:-0.50 to -0.06; p = 0.01) and in the expressive language (aMD = -0.14; CI 95%:-0.27 to -0.01; p = 0.04), fine motor skills (aMD = -0.17; CI 95%: -0.33 to -0.01; p = 0.03), gross motor (aMD = -0.22; CI 95%:-0.40 to -0.04; p = 0.02), and visual reception (aMD = -0.22; CI 95%:-0.40 to -0.05; p = 0.01) domains. Higher maternal stress was associated with poorer overall infant cognitive scores (aMD = -0.11; CI 95%:-0.20 to -0.02; p = 0.02) and in the specific domains of expressive language (aMD = -0.07; CI 95%:-0.12 to -0.01; p = 0.01), gross motor skills (aMD = -0.12; CI 95%:-0.18 to -0.05; p < 0.01) and visual reception (aMD = -0.09; CI 95%:-0.16 to -0.02; p = 0.02). Comparisons between the small number of HIV positive infants (n = 16) and the HEU infants (n = 381) showed the latter to have higher mean gross motor scores (50.3 vs. 40.6; p = 0.01). There was no evidence of difference by HIV status in the other MSEL domains or overall mean cognitive scores. Our findings demonstrate the association between maternal mood and stress levels and child cognitive functioning, particularly in expressive language and visual reception development. Although cross sectional data cannot shed light on the direction of this association, the study suggests that interventions to address maternal stress and depression symptoms may prove to be beneficial.