This research aims to recalibrate nuclear deterrence theory by replacing "outcome-based" utility functions found in classic models (Shubik) with "language-based" utility functions (Capraro et al., 2024) tailored for theocratic autocracies. By quantifying ideological sentiment within predictive mathematical engines (Bueno de Mesquita), the study provides a more accurate framework for predicting escalation in multi-polar, ideologically driven nuclear environments.
References: 1 Shubik, M. (1987). Game Theory Models of Strategic Behavior and Nuclear Deterrence. Cowles Foundation Discussion Paper No. 525, Yale University. 2 Naghavi, A., & Pignataro, G. (2015). Theocracy and Resilience Against Economic Sanctions. (Journal of Comparative Economics). 3 Prinz, A. L., & Sander, C. J. Political leadership and the quality of public goods and services: Does religion matter? 4 Bueno de Mesquita, B. (2011). A New Model for Predicting Policy Choices Preliminary Tests. 5 Scholz, J. B., Calbert, G. J., & Smith, G. A. (2011). Unravelling Bueno de Mesquita’s Group Decision Model. 6 De Mesquita, B. B. (2010). The Predictioneer's Game: Using the Logic of Brazen Self-Interest to See and Shape the Future. Random House.7 Seal, J. S. (2013). Game Theory, Predictive Analysis, And Iran. 8 Capraro, V., Di Paolo, R., Perc, M., & Pizziol, V. (2024). Language-based game theory in the age of artificial intelligence. Journal of the Royal Society Interface.
- Alejandro, Earl B., "The Leader's Gamble: Political and Military Decision-Making and Risk-Taking in Battle and War," SAASS thesis, 2024.
- Alejandro critiques Rational Choice Theory (RCT) and Expected Utility Theory (EUT)—the baseline of Shubik's classic models—by proving that political and military leaders do not always prefer options that maximize objective, material utility. Drawing on prospect theory, the paper explains that actors appraise risk situationally based on a cognitive "reference point" of gains and losses, meaning that highly motivated leaders will accept extreme risks and absorb high costs simply to recoup perceived losses. Furthermore, because leaders inevitably assess uncertainty based on personal convictions and subjective probability rather than objective truths, classic mathematical engines must be updated to incorporate these psychological and ideational biases.
- Bergin, Capt. Conor T., "Beyond Brinksmanship: How Evolving Nuclear Deterrence Endangers Strategic Stability," AFGC thesis, 2025.
- Explores how the transition from Cold War bipolar simplicity to a highly complex, multipolar nuclear landscape challenges the foundational logic of Mutually Assured Destruction (MAD). Bergin explains that doctrinal volatility, technological disruptions, and the introduction of additional nuclear-armed actors interact in ways that compress decision-making windows, erode mutual understanding, and dramatically increase the risk of miscalculation. Although the paper does not mathematically quantify ideological sentiment, it highlights the urgent need to balance modern capabilities with communication frameworks that can ease the compressed timelines of a fragmented strategic landscape.
- Crouch, Maj. Daniel W., "Rethinking Nuclear Crisis Management," ACSC thesis, 2022.
- This research recalibrates the standard utility functions of nuclear decision-makers by moving away from classic "rational choice" models and applying behavioral economics. Utilizing Tversky and Kahneman's Prospect Theory, the paper demonstrates that strategic decision-making under risk is heavily dictated by how a situation is framed. It argues that as a high-stakes crisis escalates toward the brink, leaders are likely to adopt a "loss framing" perspective, which significantly increases their propensity for highly risky, risk-seeking behaviors—including nuclear first-use. By highlighting how cognitive biases and emotional framing can overrule rational expected utility calculations, this research provides a more mathematically and psychologically realistic engine for predicting escalation dynamics.
- Douglas, Jamie, "The Likelihood of Multilateral Treaties, Bilateral Treaties or a Nuclear Arms Race," SAASS thesis, 2021.
- This paper explores how different types of government shape strategic decision-making and willingness to negotiate strategic nuclear balances. Drawing on Bruce Bueno de Mesquita's selectorate theory (The Logic of Political Survival), the author highlights how the size of a leader's winning coalition and selectorate dictates their focus on regime survival and national security definitions. It argues that standard rational deterrence assumptions fail to account for cultural and religious inputs—such as Russia's state-sanctioned religion—which can directly guide state choices and shift a leadership's risk-tolerance. Ultimately, this framework demonstrates that rationality and risk calculations are highly subjective and "in the eye of the beholder," necessitating a recalibration of classic expected utility models to fit autocratic or ideologically driven regimes.
- Garrison, Maj. John, "Strategic Messaging Implications for Military Reactions in the Taiwan Strait," AFGC thesis, 2025.
- By leveraging Large Language Models (LLMs) like GPT-4 to parse, clean, and score over two million words across thousands of official articles and speeches, the author establishes a structured database to track escalatory rhetoric. Garrison demonstrates how automating large-scale content analysis allows researchers and intelligence analysts to detect rhetorical shifts and thematic trends that were previously impractical to study. This data-driven, computational approach offers a robust foundation for building predictive AI models that can forecast an adversary's specific military or rhetorical reactions to diplomatic policies based on language-based sentiment rather than static outcome-based metrics.
- Isom, Joshua M., "2022 Russian-Ukrainian War: Analysis Using Three Deterrence Models," SAASS thesis, 2025.
- Isom explores how classical models fail when they ignore the unique motivations, cultural factors, and ideological resolve of an adversary. The research highlights that a challenger's decision to escalate is ultimately determined by their subjective perception of whether they can calculate and control operational risks, rather than the objective, local military balance. By synthesizing qualitative frameworks that map an opponent's subjective beliefs (such as George & Smoke's conditions) with objective, structural indicators (such as Huth & Russett's trade and arms transfer metrics), Isom shows how to build a more comprehensive and accurate predictive engine for ideologically motivated conflicts.
- McDermott, Maj. Dylan, "The Myth of Cyber Dominance," ACSC EL, 2019.
- Discusses the difficulty of calculating adversary utility and escalation thresholds in highly ambiguous, multi-actor crises. The research notes that conventional nuclear deterrence signaling is often ill-suited for asymmetric domains where conventional and non-conventional effects are deeply entangled. McDermott argues that to effectively manage escalation, military planners must move beyond flat, effects-based logics and instead design strategies tailored to the specific cognitive vulnerabilities, perceptions, and behavioral calculations of individual threat actors.
- Morris, Wade F., "Keeping Your AI on the Road: Increasing Asymmetric Advantage through Artificial Intelligence and International Agreements," SAASS thesis, 2024.
- The research argues that to make international commitments and strategic messaging effective, practitioners must move beyond traditional, retrospective assessments and instead employ big data and machine learning to capture real-time public and political sentiment. Morris illustrates how sentiment analysis functions as a critical cognitive feedback loop at the tactical and strategic levels. By systematically measuring host-nation and adversary sentiment, military planners can continuously calibrate operational strategies, demonstrating how language-based feedback can optimize and stabilize competitive instruments of power within contested environments
- Mueller, Brian C., "Prelude to War or Competition? The Strategic Implications of Deterrence through Peacetime Airpower Projection," SAASS thesis, 2024.
- Applying "emotional choice theory" (the logic of affect), Mueller demonstrates that specific emotional states—fear, anger, hope, pride, and humiliation—actively shape how state leaders appraise threats and select strategic preferences, often overriding purely logical calculations of material costs. In highly contested or ideologically driven scenarios, coercive military signals can trigger intense emotional reactions that inadvertently accelerate escalation rather than deterring conflict. To build an accurate predictive model, the author argues that strategists must abandon a purely material lens and integrate these intangible, cognitive-emotional dynamics into the adversary's decision calculus.
- Stinson, Joshua S., "Stubborn Giants: Assessing Resolve between the United States and China," SAASS thesis, 2020.
- These papers address how to recalibrate expected utility equations by incorporating Joshua Kertzer's interactionist Theory of Resolve. The author explains that traditional models struggle to predict specific crisis outcomes because they rely on simplistic, outcome-based cost-of-fighting averages. To solve this, the resolve framework replaces classic cost-benefit models with a complex interaction of national-level situational factors and leader-level dispositional variables, such as risk preferences, time preferences, and honor orientation. This interactionist model allows practitioners to map and target an opponent's specific dispositional vulnerabilities, offering a much more precise mechanism for predicting escalatory resolve during regional confrontations.
- Tatum, Lt. Col. William C., "Kaleidoscope: Three Analytical Lenses for Viewing the Past, Present and Future of Iranian Nuclear Weapons Ambitions," AWC SSP, 2019.
- This study directly addresses how to analyze and predict the nuclear strategic choices of Iran, a prominent theocratic autocracy. The author explicitly challenges the dominance of traditional, Cold War-era strategic nuclear deterrence models, arguing that regional powers conceptualize the utility of nuclear weapons differently than superpowers. Using Vipin Narang's postures of regional states, the paper predicts whether an ideologically driven regime will choose an asymmetric escalation or an assured retaliation strategy. It further utilizes Mark Bell's theory of "emboldenment" to show how the possession of nuclear weapons fundamentally recalibrates a state's strategic cost-benefit calculus, allowing them to reallocate military resources and aggressively pursue new foreign policy interests.
- Thrasher, Maj. Michael, "Doomed to Succeed? Understanding Nuclear Deterrence and Its Past and Future Roles in National Security," AFGC thesis, 2024.
- Drawing on the psychological deterrence theories of Jervis, Lebow, and Stein, Thrasher emphasizes that effective deterrence is fundamentally a cognitive problem that requires analyzing how opponents process information, rather than assuming they behave as objective utility maximizers. The paper explains that political leaders' prior beliefs, confirmation biases within government decision-making structures, and the non-attributional nature of modern conflicts introduce severe strategic uncertainty. Consequently, predicting escalation in multipolar, ideologically driven environments requires a major paradigm shift—moving away from rigid, material cost calculations and toward a deep, qualitative understanding of how an adversary's unique worldview, cultural biases, and domestic political constraints filter strategic communication.
- Unruh, Maj. Michael Y., "21st Century Coercion: Growing Deterrence Doctrine, Policy and Strategy into the 2020s and Beyond," AFGC thesis, 2020.
- Provides critical context on the inherent flaws of using uniform rational-actor models to predict foreign behaviors. Unruh argues that a prominent historical fallacy of Cold War deterrence was the assumption that Soviet leaders utilized a similar decision-making calculus to Western leaders. While the study does not model theocratic autocracies using language-based utility functions, it emphasizes that modern 21st-century coercion cannot rely on mirror-imaging; instead, military planners must build more accurate predictive models by deeply analyzing an adversary's unique stated objectives, strategic culture, and domestic messaging.
- Wade, Lt. Col. Jonathan H., "Existential Threat and Religious Violence Origins, Predictability and Mitigation," AWC SSP, 2018.
- This work provides a methodology for predicting escalation and violence among highly ideologically driven groups, arguing that classic, purely materialistic cost-benefit models are insufficient. Utilizing Terror Management Theory (TMT), the paper models how an existential threat to a religious worldview serves as a powerful motivator that drives collectives to shift from peaceful status quo to violent, defensive escalation. To quantify this ideological sentiment, the author presents a threat-violence continuum as a predictive tool to help strategic planners evaluate potential outbreaks of conflict and anticipate behavioral shifts. This non-kinetic framework focuses on narrative-based and soft-power options to reduce worldview-related anxiety, offering a structured way to predict and de-escalate tension where conventional deterrence fails.