Answers are grounded in this paper and cite their sources.
Networks of interacting oscillators can move together to much the same degree while behaving very differently over time. This distinction matters because average synchronization alone may overlook how long coordinated activity persists along a network’s connections. We investigated this hidden temporal memory using a noisy oscillator model and a measure comparing the instantaneous phase agreement of connected and unconnected node pairs. In 30 independent pairs of 128-node small-world networks, changing the connections while preserving every node’s degree reduced the measure’s correlation time from 2.178 to 1.159 time units, despite nearly identical average global synchronization (0.395 and 0.394). The difference remained, although smaller, after accounting for fluctuations in global synchronization. We then showed why network structure matters: a spectral description identifies the collective modes that contribute to memory and predicts their relaxation near synchronized states. Further simulations linked memory to dynamical modes in heterogeneous networks and demonstrated the importance of the sequence of interactions when synchronization is incomplete. These findings show that network architecture shapes not only how strongly a system coordinates, but also how long its coordination patterns persist. This perspective may inform studies of neural rhythms, engineered oscillator arrays and other interconnected systems.
No takes yet. Share an insight, caveat, or question.
Yousef Jamali (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: