Trust Me, I Can Convince You: The Contextualized Argument Appraisal Framework

TL;DR

The paper addresses how people emotionally respond to arguments not only from content but also from contextual appraisal of impacts on the self. It introduces the Contextualized Argument Appraisal Framework (CAAF) and a role-playing annotation setup to model the interplay between sender, receiver, and argument, culminating in the ContArgA corpus of 4000 annotations across 800 arguments. Key findings show that convincingness correlates positively with positive emotions (e.g., trust, relief, pride, joy) and negatively with negative emotions (e.g., anger, disgust), with argument familiarity emerging as a key predictor of appraisal and emotion. The work provides a foundation for context-sensitive, emotionally aware computational models of argument persuasiveness and suggests practical applications in argument mining, moderation, and disinformation research, while releasing the ContArgA data for future study.

Abstract

Emotions that somebody develops based on an argument do not only depend on the argument itself - they are also influenced by a subjective evaluation of the argument's potential impact on the self. For instance, an argument to ban plastic bottles might cause fear of losing a job for a bottle industry worker, which lowers the convincingness - presumably independent of its content. While binary emotionality of arguments has been studied, such cognitive appraisal models have only been proposed in other subtasks of emotion analysis, but not in the context of arguments and their convincingness. To fill this research gap, we propose the Contextualized Argument Appraisal Framework to model the interplay between the sender, receiver, and argument. We adapt established appraisal models from psychology to argument mining, including argument pleasantness, familiarity, response urgency, and expected effort, as well as convincingness variables. To evaluate the framework and pave the way for computational modeling, we develop a novel role-playing-based annotation setup, mimicking real-world exposure to arguments. Participants 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 ask them to disclose the same variables for their perception of the argument sender. The analysis of the resulting ContArgA corpus of 4000 annotations reveals that convincingness is positively correlated with positive emotions (e.g., trust) and negatively correlated with negative emotions (e.g., anger). The appraisal variables particularly point to the importance of the annotator's familiarity with the argument.

Figures (9)

• Figure 1: Illustration of the Contextualized Argument Appraisal Framework. The perception of the argument, context and sender are cognitively evaluated and assessed as having a negative consequence for the receiver (e.g., losing a job if plastic bottles were banned).
• Figure 2: Flowchart of the annotation workflow. Yellow fields indicate information collected about the receiver (the annotator), green fields indicate argument and scenario display and blue fields indicate argument specific annotation.
• Figure 3: Emotion Intensities and Convincingness.
• Figure 4: Analysis of discrete emotion categories and appraisal dimensions in arguments. Each row is an appraisal, each column is an emotion, and each cell is the average appraisal value for that emotion.
• Figure 5: Description of the role-playing scenario in our annotation study.