Category: Network Sci.

Summary: Testing whether moderately negative degree-threshold correlation widens the cascade window in spatial threshold dynamics on random geometric graphs.

Threshold cascades on spatial networks depend both on who is connected and on how easily each node activates. This experiment asks whether making high-degree nodes slightly easier to activate widens the global cascade window, while pushing that anticorrelation too far instead traps activation around hub neighborhoods.

The simulation runs threshold cascades on random geometric graphs while tuning the correlation between node degree and activation threshold. It tracks whether moderate anticorrelation helps hubs recruit distant regions without making the low-degree periphery too hard to ignite.

That separates a subtle design question that is often obscured in aggregate cascade studies. The result targets whether the right amount of hub facilitation changes not just cascade size, but the size of the parameter region where broad cascades are possible.

Method: Repeated threshold-cascade simulations on random geometric graphs with tunable degree-threshold correlation.

What is measured: Cascade-window width, cascade size, dependence on degree-threshold correlation, hub-seeding effects, and evidence for an interior optimum.