Sparse research suggests that children's social information processing has links not just with aggressive behavior but also with children's prosocial behavior (e.g., Nelson & Crick, 1999). However, the past work that has been done has not been longitudinal, so the direction of links between social information processing and prosocial behavior remains unclear. In this study, we used data from the National Institute of Child Health and Human Development Study of Early Child Care and Youth Development (NICHD Early Child Care Research Network, 2010) to examine longitudinally the links between prosocial as well as aggressive behaviors and social information processing. Children completed multiple assessments of social information processing (including attribution biases and strategy response selection) from the 3rd to 5th grades. Mothers and teachers completed measures of children's prosocial and aggressive behavior from the 3rd to 6th grades. Overall, the findings demonstrated that some of the links between social information processing and social behavior are bidirectional but that the direction of effects depends on when such links were assessed. At Grade 3, it was mostly children's social behavior that predicted social information processing. At Grades 4 and 5, however, social information processing predicted children's social behavior.