Persona-E$^2$: A Human-Grounded Dataset for Personality-Shaped Emotional Responses to Textual Events

TL;DR

This paper introduces Persona-E$^2$, a large-scale dataset linking personality traits to emotional responses to textual events, highlighting the importance of personality-aware models in affective computing.

Contribution

The creation of Persona-E$^2$, a dataset grounded in MBTI and Big Five traits, enabling better understanding of personality-influenced emotional appraisals in text.

Findings

State-of-the-art LLMs struggle with emotional appraisal shifts.

Personality information improves model understanding of emotional responses.

Big Five traits help reduce 'personality illusion' in LLMs.

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.'