nucleation

LIT-960: Symmetries, Cluster Synchronization, and Isolated Desynchronization in Complex Networks

2 min read

Properties5
StatusActive — read 2026-10-14 (NOTE-775) in full; worth reading as the general account of cluster synchronization for identical units on a symmetric graph: the symmetry orbits are the clusters that can synchronize, irreducible representations separate each cluster's stability, and a cluster moved by its own factor of the group can desynchronize alone, as an electro-optic experiment showed. Read it knowing that symmetry is sufficient, not necessary, for clusters, that nonidentical units and weighted or directed coupling are not worked through, and that it says nothing about external drive. Its use on QUESTION-063 as a coupling signature is the record's inference, held in THEORY-276.
Version2
Tagsnetwork-science, complex-systems
Cited as “paper” byNOTE-775 — Pecora et al. — Cluster synchronization and isolated desynchronization
Cited as “source” byTHEORY-276 — In a network of identical units with identical couplings, the orbits of the graph's symmetry group are synchronization clusters, and a cluster moved only by its own factor of the group can lose synchrony while the others keep it

Graph View

Louis M. Pecora, Francesco Sorrentino, Aaron M. Hagerstrom, Thomas E. Murphy and Rajarshi Roy (2014), Nature Communications 5, 4079.

Key takeaways

From the reading, NOTE-775.

  • The wiring fixes which clusters can synchronize. For ẋᵢ = F(xᵢ) + σ Σⱼ Aᵢⱼ H(xⱼ), with identical units and a symmetric 0/1 adjacency matrix, each symmetry of the graph commutes with A. So the orbits of the symmetry group can move identically.
  • The dynamics fix which do. The IRR transformation makes TAT⁻¹ block-diagonal. Each cluster is stable where its own transverse block’s maximum Lyapunov exponent is negative, for any node dynamics.
  • Isolated desynchronization.
    • If the group splits into commuting factors, nodes outside a cluster see the same summed input from it whether or not it is synchronized.
    • So it can desynchronize alone.
    • Clusters moved by one non-decomposable factor are “intertwined” and desynchronize together.
  • Observed. In an 11-node, 32-symmetry electro-optic network, the four-node cluster split while the others held (β = 1.4π). Two intertwined clusters lost synchrony together. Simulated exponents matched both.
  • Common. Almost all of 10,000 random, Barabási–Albert and bipartite 100-node graphs of each kind had several nontrivial clusters and several factors. So did two power grids.
  • Limits stated in the paper. Symmetry is sufficient, not necessary: clusters also arise from matched input and output couplings without symmetry. Nonidentical units and weighted or directed coupling “can all be handled in much the same way”, but are not worked through.

Standing in the record

Filed on 2026-10-12 for QUESTION-063 and read on 2026-10-14 in NOTE-775. Its finding is filed as THEORY-276. The registration said its clusters are “a coupling signature that a shared external drive does not predict”. The paper never considers external drive. The inference is the record’s, and it is sound only within the paper’s class: identical units on a known graph. THEORY-276 states it in that form. It also bears on CLAIM-172 and CLAIM-157. A mutually coupled network yields several synchronized sub-collectives whose members need not be linked, and their invariance comes from equal input, not from weak coupling across a block’s edge.

Backlinks