Abstract
<p>Cyber conflict has become central to international politics, yet traditional strategic frameworks, such as deterrence, struggle to explain competition below the threshold of armed conflict. Cyber Persistence Theory conceptualizes cyberspace as a distinct strategic environment in which security depends on seizing and sustaining initiative through persistent engagement. Yet the theory faces an empirical deadlock: cyber operations are largely classified and intentionally difficult to observe, making initiative challenging to trace empirically; in addition, available measures tend to capture isolated incidents rather than cumulative strategic effects. To address this problem, this study treats cyberspace as a complex adaptive system and develops a nested agent-based model of interstate cyber competition in a network-of-networks characterized by ever-present vulnerabilities, limited resources, and partial visibility. The model bridges micro-level interactions over exploitation and recovery with macro-level outcomes in strategic advantage. The paper formalizes Cyber Persistence Theory by operationalizing initiative as control, shifting the analysis from one-off compromise events to how control is accumulated and retained over time. This enables analysis of mechanisms through which persistent engagement produces compounding gains. The simulations show how hybrid persistent-engagement strategies can outperform pure offense or defense by combining recovery, foreign access, operational tempo, and visibility advantages. The model therefore provides a transparent, mechanism-based test of Cyber Persistence Theory's core logic, showing how persistent engagement can convert maintained initiative into cumulative strategic advantage. More broadly, the study demonstrates the value of agent-based modeling for analyzing strategic tradeoffs in cyber conflict, where direct empirical evidence is scarce and likely to remain so.</p>