Abstract
Modularity is a ubiquitous topological feature of structural brain networks at various scales. While a variety of potential mechanisms have been proposed, the fundamental principles by which modularity emerges in neural networks remain elusive. We tackle this question with a plasticity model of neural networks derived from a purely topological perspective. Our topological reinforcement model acts enhancing the topological overlap between nodes, iteratively connecting a randomly selected node to a non-neighbor with the highest topological overlap, while pruning another network link at random. This rule reliably evolves synthetic random networks toward a modular architecture. Such final modular structure reflects initial ‘proto-modules’, thus allowing to predict the modules of the evolved graph. Subsequently, we show that this topological selection principle might be biologically implemented as a Hebbian rule. Concretely, we explore a simple model of excitable dynamics, where the plasticity rule acts based on the functional connectivity between nodes represented by co-activations. Results produced by the activity-based model are consistent with the ones from the purely topological rule, showing a consistent final network configuration. Our findings suggest that the selective reinforcement of topological overlap may be a fundamental mechanism by which brain networks evolve toward modular structure.