Trust Me, I Can Convince You: The Contextualized Argument Appraisal Framework (2509.17844v1)

Abstract: Emotions, which influence how convincing an argument is, are developed in context of the self and sender, and therefore require modeling the cognitive evaluation process. While binary emotionality has been studied in argument mining, and the cognitive appraisal has been modeled in general emotion analysis, these fields have not been brought together yet. We therefore propose the Contextualized Argument Appraisal Framework that contextualizes the interplay between the sender, receiver, and argument. It includes emotion labels, appraisals, such as argument familiarity, response urgency, and expected effort, as well as convincingness variables. To evaluate the framework and pave the way to computational modeling, we perform a study in a role-playing scenario, mimicking real-world exposure to arguments, asking participants to disclose their emotion, explain the main cause, the argument appraisal, and the perceived convincingness. To consider the subjective nature of such annotations, we also collect demographic data and personality traits of both the participants and the perceived sender of the argument. The analysis of the resulting corpus of 800 arguments, each annotated by 5 participants, reveals that convincingness is positively correlated with positive emotions (e.g., trust) and negatively correlated with negative emotions (e.g., anger). The appraisal variables disclose the importance of the argument familiarity. For most participants, the content of the argument itself is the primary driver of the emotional response.