We examined the longitudinal relationship between IQ and fine particulate matter (\2.5lm aerodynamic diameters; PM2.5) exposure in urban-dwelling children, using prospective longitudinal data from the USC Twin Study of Risk Factors for Antisocial Behavior (RFAB; Baker et al. 2013). Residential addresses were collected via selfreports. Verbal and Performance IQ during childhood (age 9–10) and young adulthood (age 19–20) were evaluated by the Wechsler Abbreviated Intelligence Scale (Wechsler, 1999) using four subtests: VIQ=Vocabulary Similarities; PIQ=Block Design Matrices. Based on residential addresses and spatiotemporal generalized additive model of local monitoring data for PM2.5, we estimated 1-year average exposure before each assessment. A three-level mixed effects model regressing IQ scores at each assessment on time-varying air pollution exposures, accounting for both within-family (random intercepts) and within-individual (random slopes) was used. PM2.5 exposure had significant adverse effects on PIQ (95 % CI of b:-7.29 to-1.01, p\.05) but not VIQ (95 % CI of b:-4.50 to-1.96). Adverse effects of PM2.5 exposure remained significant after adjusting for age, family SES, sex, race/ethnicity, parental cognitive abilities, neighborhood SES, neighborhood quality and neighborhood greenness; the association was still significant after further adjusting for traffic distance (300 m), temperature, humidity and annual NOx. PM2.5 exposure confers stronger adverse effects on PIQ in low SES families, males, and during pre-adolescence. Our findings reveal social disparities and sexual dimorphism in the adverse PM2.5 exposure effects on PIQ. Baker, L., Tuvblad, C., Wang, P., Gomez, K., Bezdjian, S., Niv, S., & Raine, A. (2013). The Southern California Twin Register at the University of Southern California: III. Twin Research and Human Genetics, 16(1), 336–343; Wechsler, D. (1999). Wechsler Abbreviated Scale of Intelligence (WASI). San Antonio, Texas: Harcourt Assessment.