A framework for the local information dynamics of distributed computation in complex systems

The nature of distributed computation has often been described in terms of the component operations of universal computation: information storage, transfer and modification. We review the first complete framework that quantifies each of these individual information dynamics on a local scale within a system, and describes the manner in which they interact to create non-trivial computation where “the whole is greater than the sum of the parts”. We describe the application of the framework to cellular automata, a simple yet powerful model of distributed computation. This is an important application, because the framework is the first to provide quantitative evidence for several important conjectures about distributed computation in cellular automata: that blinkers embody information storage, particles are information transfer agents, and particle collisions are information modification events. The framework is also shown to contrast the computations conducted by several well-known cellular automata, highlighting the importance of information coherence in complex computation. The results reviewed here provide important quantitative insights into the fundamental nature of distributed computation and the dynamics of complex systems, as well as impetus for the framework to be applied to the analysis and design of other systems.

💡 Research Summary

The paper presents a comprehensive information‑theoretic framework that quantifies the three elementary operations underlying distributed computation—information storage, information transfer, and information modification—on a local (spatiotemporal) scale. Building on earlier concepts such as active information storage (AIS) and transfer entropy (TE), the authors introduce a novel metric for information modification that captures the synergistic creation of new information when multiple information streams intersect. By evaluating AIS, TE, and the modification measure simultaneously at every cell and time step, the framework reveals how these dynamics interact to produce emergent computation that cannot be understood by looking at components in isolation.

The authors apply the framework to cellular automata (CA), a paradigmatic model of distributed computation. Using well‑studied rules (e.g., Rule 110, Rule 54, Rule 30), they demonstrate that static periodic structures (“blinkers”) exhibit high AIS, confirming that they act as memory reservoirs. Mobile structures (“particles”) show elevated TE, indicating that they are the primary carriers of information across the lattice. Crucially, at particle collision sites both TE drops sharply and the information‑modification metric spikes, providing quantitative evidence that collisions are genuine information‑modifying events. This empirical validation settles long‑standing conjectures that blinkers store, particles transfer, and collisions modify information.

Beyond confirming these conjectures, the framework distinguishes the computational character of different CA rules. Chaotic rules such as Rule 30 display low, spatially uniform AIS and TE, and a diffuse modification profile, reflecting a lack of coherent information flow and thus limited capacity for structured computation. In contrast, complex rules like Rule 110 and Rule 54 exhibit localized pockets of high AIS (memory) and TE (communication) that are linked by sharp peaks of modification at collision points. This pattern of “information coherence”—the alignment of storage, transfer, and modification in space and time—correlates with the ability of the system to support non‑trivial, self‑organized computation, embodying the principle that “the whole is greater than the sum of its parts.”

The authors argue that the framework’s locality is its key advantage over traditional global entropy‑based measures. By exposing where and when information is stored, moved, or transformed, researchers can trace the causal architecture of complex dynamics, identify functional modules, and even guide the design of engineered systems. They suggest that the same methodology could be transferred to neural networks (synaptic plasticity as storage, spikes as transfer, nonlinear integration as modification), gene regulatory networks, social systems, and other domains where distributed processing occurs.

In summary, the paper delivers the first complete, quantitative toolkit for dissecting distributed computation at the microscopic level. Its application to cellular automata not only validates longstanding hypotheses about the roles of blinkers, particles, and collisions but also uncovers the central importance of information coherence in fostering complex, emergent behavior. The work opens a pathway for systematic analysis and purposeful design of complex information‑processing systems across physics, biology, and engineering.