Abstract
Nature is a blossoming of regular structures, signature of self-organization of the underlying microscopic interacting agents. Turing theory of pattern formation is one of the most studied mechanisms to address such phenomena and has been applied to a widespread gallery of disciplines. Turing himself used a spatial discretization of the hosting support to eventually deal with a set of ODEs. Such an idea contained the seeds of the theory on discrete support, which has been fully acknowledged with the birth of network science in the early 2000s. This approach allows us to tackle several settings not displaying a trivial continuous embedding, such as multiplex, temporal networks, and, recently, higher-order structures. This line of research has been mostly confined within the network science community, despite its inherent potential to transcend the conventional boundaries of the PDE-based approach to Turing patterns. Moreover, network topology allows for novel dynamics to be generated via a universal formalism that can be readily extended to account for higher-order structures. The interplay between continuous and discrete settings can pave the way for further developments in the field.
Introduction
In 1952, Alan Turing proposed that a stable system of reacting and diffusing chemicals could spontaneously break symmetry to form patterns—such as spots or stripes—if one species diffuses significantly faster than the other. While groundbreaking, classical Turing theory relies on continuous space and partial differential equations (PDEs), which often impose restrictive conditions that do not always align with the diversity of patterns observed in nature.
The real world is often intrinsically discrete, ranging from neurons connected by axons to species moving between ecological patches. This paper reviews the generalization of Turing’s ideas to discrete topologies. By transitioning from simple grids to complex networks and "higher-order" structures (where interactions occur in groups rather than just pairs), researchers can model more realistic systems and discover novel mechanisms for pattern emergence.
Theoretical Foundations
The study of dynamics on discrete supports necessitates a synthesis of several mathematical frameworks. Central to this is the Graph Laplacian, which replaces the continuous diffusion operator ($\nabla^2$) with a matrix representation of a network's connectivity.
Linear Stability Analysis is adapted for the eigenvalues of the Graph Laplacian to determine if small perturbations will grow into patterns. The dispersion relation $\lambda(\Lambda)$ characterizes the growth rate of different modes. Furthermore, the Master Stability Function (MSF), borrowed from synchronization theory, allows for the analysis of Turing-like instabilities in oscillatory or chaotic systems.
Topology-Driven Instabilities
A key insight of this review is the role of network architecture in catalyzing instabilities. In continuous space, diffusion is typically symmetric. However, on directed networks, the resulting "non-normal" dynamics significantly lower the threshold for pattern formation.
Crucially, directedness allows for oscillatory patterns in systems with only two species—a phenomenon that requires at least three species in the classical continuous framework. The imaginary part of the Laplacian spectrum in directed networks acts as a "boost" for instability, effectively detaching the discrete dispersion relation from its continuous counterpart. This "topology-driven path" highlights how discrete structures are not merely approximations but unique drivers of self-organization.
Higher-Order Structures and Topological Signals
Moving beyond pairwise links, the paper explores higher-order structures like hypergraphs and simplicial complexes. These frameworks account for multi-body interactions (e.g., three species interacting simultaneously), which can act as a "tuning knob" to either trigger or suppress instabilities. While pairwise interactions might fail to produce a pattern, adding 3-body interactions provides a new dimension of control.
Additionally, the use of the Dirac operator introduces a new class of "Dirac-induced" patterns. Traditionally, patterns are viewed as concentrations at specific nodes. This approach highlights "topological signals," where patterns manifest as specific distributions of flows across links (1-simplices) or triangles (2-simplices) rather than just values at nodes (0-simplices).
Conclusion and Future Directions
While the theoretical framework for discrete Turing patterns is robust, several challenges remain. Predicting the exact shape of a pattern based solely on complex network architecture is still difficult; a comprehensive "dictionary" between network metrics and pattern motifs is needed. Future work will likely explore "pinning control" to steer pattern formation, the extension of these models to directed hypergraphs, and the nascent study of Turing instabilities in quantum systems. The interplay between continuous and discrete settings continues to pave the way for further developments in the field.
Summary updated