Recent advances in Magnetic Resonance Imaging are allowing neuroscientists to gain critical insights into the neural networks mediating a variety of cognitive processes. Neural networks are sets of anatomically and functionally interconnected regions and are the origin of cognitive pro- cesses. In this work, we studied the relationship between structure and function in these networks, through the two fundamental principles of brain function, namely segregation and integration, and the concept of integrity. First, we focused on the anatomical segregation of basal ganglia and its functional interpretation. Then we examined the principle of integration, from a descriptive way to an explanatory approach, by using the spatial autoregressive model. Finally, we studied the structu- ral integrity of the brain in the presence of mild traumatic brain injury associated neurobehavioral deficits. We demonstrated the existence of an anatomical substrate underlying the functional net- works. Our results suggest that the anatomical structure of brain networks is a complex substrate optimizing cognitives processes. Moreover, loss of structural integrity in mild traumatic brain injury affects behavior and cognitive performances. This demonstrates that brain function, mediated by neural networks, is intimately related to their anatomical structure.