4 November 2025 13h30-14h00
Vincent Joassin (COST)
Can threat perceptions embedded in political narratives serve as a determinant for assessing military expenditure? While existing research often relies on objective indicators—such as geographic proximity or the military spending of adversaries—to measure threats, these approaches frequently overlook the subjective perceptions that ultimately shape policy decisions. The political science literature has long recognized threat perception as a key driver of military spending; however, it still lacks a consistent and scalable approach to assess it systematically across time and diverse geopolitical contexts. This research project seeks to address this gap by developing a methodology to estimate national-level threat perceptions which can inform both quantitative and qualitative analyses. Specifically, it leverages large-scale political texts and the natural language processing capabilities of large language models (LLMs) to test threat perception indicators derived from the analysis. Such approach would enable researchers to capture more of the subjective dimensions of threat perceptions, offering a new and replicable tool for studying the relationship between political narratives, threat perceptions, and defence policy decisions.
Teams Link will be communicated very soon