A study published in Nature proposes a theoretical framework to describe higher-order contagion processes in complex networks, where interactions involve groups of nodes of varying size rather than simple pairs. Such processes, common in disease spread, rumor propagation and behavior adoption, cannot be captured by classical models based solely on binary links.

The work addresses a known limitation of network theory: most formalisms describe contagion through one-to-one interactions, whereas in real systems collective events —gatherings, classrooms, markets or digital platforms— generate correlations that substantially alter propagation thresholds. By incorporating hypergraphs and simplicial complexes, the authors unify in a single treatment the contributions of different interaction scales.

The method combines tools from algebraic topology with stochastic process theory to derive evolution equations that simultaneously integrate two-, three- and higher-order interactions. The formulation identifies how each interaction scale modifies the critical point of the epidemic transition and the diffusion speed, showing that group effects can lower the contagion threshold compared with pair-based predictions.

The results have implications for designing containment strategies in epidemiology, managing information in social networks and understanding collective phenomena in biological and technological systems. The framework also provides a basis for future simulations and for the empirical analysis of high-resolution contact data, where group interactions are the norm rather than the exception.