This article introduces the MASS model—analyzing audio and speech structure—to systematically study auditory cues in political communication.
Political Speech as a Stochastic Process: We model political dialogue not just through text, but by incorporating vocal tone as influenced by fixed and evolving variables.
Method: Supreme Court Oral Arguments Analysis: Applying our approach to Court proceedings demonstrates that questioning reflects genuine engagement with arguments rather than strategic manipulation.
Key Finding on Vocal Tone: Justice skepticism conveyed through tone is largely missed by traditional word-based models.
Practical Tool: The R package "communication" offers accessible software for audio analysis, extending beyond the proposed model's applications.
Our work bridges communication studies and political science by quantifying previously unexamined dimensions of interaction.
