Emotionally Charged, Logically Blurred: AI-driven Emotional Framing Impairs Human Fallacy Detection
This addresses the problem of AI-driven emotional manipulation in public communication, particularly for audiences vulnerable to fallacious arguments, though it is incremental as it builds on existing research on fallacies and emotional appeals.
The study investigated how AI-generated emotional framing affects human ability to detect logical fallacies, finding that LLM-driven emotional appeals reduced human fallacy detection by 14.5% on average and increased convincingness for emotions like fear and sadness.
Logical fallacies are common in public communication and can mislead audiences; fallacious arguments may still appear convincing despite lacking soundness, because convincingness is inherently subjective. We present the first computational study of how emotional framing interacts with fallacies and convincingness, using large language models (LLMs) to systematically change emotional appeals in fallacious arguments. We benchmark eight LLMs on injecting emotional appeal into fallacious arguments while preserving their logical structures, then use the best models to generate stimuli for a human study. Our results show that LLM-driven emotional framing reduces human fallacy detection in F1 by 14.5% on average. Humans perform better in fallacy detection when perceiving enjoyment than fear or sadness, and these three emotions also correlate with significantly higher convincingness compared to neutral or other emotion states. Our work has implications for AI-driven emotional manipulation in the context of fallacious argumentation.