How to suppress undesired synchronization

TL;DR

This paper investigates strategies to suppress undesired synchronization in complex systems, demonstrating that local information-based contrarians, especially at highly connected nodes, effectively mitigate synchronization in both artificial and real networks.

Contribution

It introduces and compares various suppression strategies, highlighting the effectiveness of local information-based contrarians at highly connected nodes for mitigating undesired synchronization.

Findings

Local information-based contrarians are most effective.

Contrarians at highly connected nodes improve mitigation.

Results are consistent across artificial and real networks.

Abstract

It is delightful to observe the emergence of synchronization in the blinking of fireflies to attract partners and preys. Other charming examples of synchronization can also be found in a wide range of phenomena such as, e.g., neurons firing, lasers cascades, chemical reactions, and opinion formation. However, in many situations the formation of a coherent state is not pleasant and should be mitigated. For example, the onset of synchronization can be the root of epileptic seizures, traffic congestion in communication networks, and the collapse of constructions. Here we propose the use of contrarians to suppress undesired synchronization. We perform a comparative study of different strategies, either requiring local or total knowledge of the system, and show that the most efficient one solely requires local information. Our results also reveal that, even when the distribution of neighboring interactions is narrow, significant improvement in mitigation is observed when contrarians sit at the highly connected elements. The same qualitative results are obtained for artificially generated networks as well as two real ones, namely, the Routers of the Internet and a neuronal network.