Objective: To investigate: 1) prospective associations between media exposure (television viewing, computers, and electronic games) at 2 years and self-regulation at 4 and 6 years, and 2) bidirectional associations between media exposure and self-regulation at 4 and 6 years. We hypothesized that media exposure and self-regulation would show a negative prospective association and subsequent bidirectional inverse associations. Methods: Data from the nationally-representative Longitudinal Study of Australian Children when children were aged 2 years (n = 2786) and 4/6 years (n = 3527) were used. Primary caregivers reported children's weekly electronic media exposure. A composite measure of self-regulation was computed from caregiver-, teacher-, and observer-report data. Associations were examined using linear regression and cross-lagged panel models, accounting for covariates. Results: Lower television viewing and total media exposure at 2 years were associated with higher self-regulation at 4 years (both β = −0.02; 95% confidence interval [CI], −0.03 to −0.01). Lower self-regulation at 4 years was also significantly associated with higher television viewing (β = −0.15; 95% CI, −0.21 to −0.08), electronic game use (β = −0.05; 95% CI, −0.09 to −0.01), and total media exposure (β = −0.19; 95% CI, −0.29 to −0.09) at 6 years. However, media exposure at 4 years was not associated with self-regulation at 6 years. Conclusions: Although media exposure duration at 2 years was associated with later self-regulation, and self-regulation at 4 years was associated with later media exposure, associations were of small magnitude. More research is needed to examine content quality, social context, and mobile media use and child self-regulation.