Mitigating cascades in coevolving networks with targeted rewiring

We investigate the coevolution of network structure and opinion dynamics by integrating a threshold-based complex contagion model with a target rewiring mechanism. In contrast to previous models that allow all nonadopting nodes to rewire indiscriminately, our framework introduces the concept of superspreader nonadopting nodes that are particularly instrumental in spreading the adoption. Only these nodes rewire their connections away from adopting neighbors to randomly chosen nonadopting nodes. Through mean-field theoretical analysis and numerical simulations, we show that this targeted rewiring strategy efficiently contains adoption cascades while significantly reducing the overall number of rewiring operations required. The presence of superspreaders naturally localizes structural adaptation to critical points in the network, leading to a more economical and stable network evolution. Our results reveal that this mechanism not only substantially reduces the cost of rewiring but also causes minimal structural change, making the system more efficient and realistic in terms of intervention cost and rewiring burden.