Persona-E$^2$: A Human-Grounded Dataset for Personality-Shaped Emotional Responses to Textual Events
career value
216K/year
🤖 AI Summary
This study addresses a critical limitation in existing affective computing approaches, which typically treat emotion as a static property of text while overlooking the diverse emotional responses elicited by individual differences in personality. To bridge this gap, the authors construct a large-scale, manually annotated dataset that integrates both MBTI and Big Five personality traits across multiple domains—including news articles, social media posts, and personal narratives—providing the first empirical resource capturing real human emotional reactions to identical stimuli under varying personality profiles. Experimental results demonstrate that current large language models struggle to accurately simulate personality-driven affective dynamics; however, explicitly incorporating Big Five personality information significantly enhances their capacity to model personalized emotional responses and effectively mitigates the “personality hallucination” problem commonly observed in role-playing scenarios.
Application Category
📝 Abstract
Most affective computing research treats emotion as a static property of text, focusing on the writer's sentiment while overlooking the reader's perspective. This approach ignores how individual personalities lead to diverse emotional appraisals of the same event. Although role-playing Large Language Models (LLMs) attempt to simulate such nuanced reactions, they often suffer from"personality illusion''-- relying on surface-level stereotypes rather than authentic cognitive logic. A critical bottleneck is the absence of ground-truth human data to link personality traits to emotional shifts. To bridge the gap, we introduce Persona-E$^2$ (Persona-Event2Emotion), a large-scale dataset grounded in annotated MBTI and Big Five traits to capture reader-based emotional variations across news, social media, and life narratives. Extensive experiments reveal that state-of-the-art LLMs struggle to capture precise appraisal shifts, particularly in social media domains. Crucially, we find that personality information significantly improves comprehension, with the Big Five traits alleviating"personality illusion.'
Problem
Research questions and friction points this paper is trying to address.
affective computing
personality
emotional appraisal
reader perspective
personality illusion
Innovation
Methods, ideas, or system contributions that make the work stand out.
personality-grounded emotion
emotional appraisal
Large Language Models
personality illusion
affective computing
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
