Convergent information flows explain recurring firing patterns in cerebral cortex.

Cortical population events, short-lived patterns of neuronal activity that recur with consistency, are central to sensorimotor coordination. These reproducible firing patterns are often attributed to attractor dynamics, supported by strong mutual connectivity. However, by using multimodal datasets-including two-photon imaging, electrophysiology and electron microscopy-we show that these reproducible patterns do not involve strongly interconnected neurons. Instead, we show that cortical networks exhibit hierarchical modularity, with core neurons serving as high-information-flow nodes at module interfaces. These cores funnel activity but lack the structural signatures of pattern-completion units that are typically found in attractor networks. Using computational models, we find that distance-dependent connectivity is necessary and sufficient to produce the modularity and transient reproducible events observed in cortex. Our findings suggest that cortical networks are preconfigured to support sensorimotor coordination. This work redefines the structural and dynamical basis of cortical activity, with a focus on the relationship between modular structure and function.