Gender Attribution in Causal Beliefs

2026-08-22

AI Summary

研究使用Twitter数据和语言模型分析了性别归因在因果关系表达中的差异，揭示了女性常被关联负面情绪及人际关系问题，而男性则更多与正面情绪及结构性影响相关。

Abstract

For centuries, women have been cast as the source of harm in public narratives, from witch hunts in early modern Europe to contemporary stereotypes about emotional instability. These cultural patterns reflect enduring biases in how people attribute causality and assign blame, often portraying women as agents of disruption and men as figures of rational authority. In this study, we examine how such gendered causal attributions appear in everyday language. Leveraging three complete 24-hour datasets of all English-language posts on Twitter, and using language models, we extract cause-and-effect relationship pairs and identify gendered attribution of causal agents. We then analyze how gender attribution relates to sentiment, the kinds of effects invoked, and the diffusion of posts through the social networks. Our findings reveal that female-attributed causes are more often associated with negative sentiment and emotional or relational outcomes, whereas male-attributed causes are more frequently linked to positive sentiment and abstract, structural effects. Moreover, male-attributed narratives spread more widely across communities. These results suggest that longstanding gender stereotypes continue to appear in how people express and amplify causal narratives in public discourse, in decentralized, high-velocity environments like social media.

Problem

• Gender Attribution
• Causal Beliefs
• Social Media
• Sentiment Analysis
• Narrative Spread

Innovation

• gendered causal attributions
• language models
• social networks
• sentiment analysis