Negative attributional style has been associated with depressive symptoms in children. Yet, it is unclear whether these cognitive biases reflect inherited characteristics of the broader depressive phenotype or are a product of children's environments. While existing data in adolescents show that negative attributions reflect a genetic predisposition, elevating depressive responses to stress, other data suggest that negative attributions in children are more likely to reflect early environmental experiences on symptoms. Here, we assess the degree to which negative attributional style and depressive symptoms arise from common genetic, shared and non-shared environmental influences in childhood. Monozygotic and dizygotic twins reported on attributional style and depressive symptoms at age 8 (n = 300 pairs) and at age 10 (n = 250 pairs). Two multivariate models with varying assumptions on the nature of the relationship between negative attributions and depressive symptoms within and across time were fit to the data. The Common Pathway model provided better fit than the Cholesky decomposition. A common, latent factor influenced both attributional style and depressive symptoms at both time-points in children. The only significant influences on this factor were shared and non-shared aspects of the environment. Placing the present findings with those of adolescents suggests possible developmental differences in the relationship between attributional style and depressive symptoms. (Contains 2 tables and 2 figures.)