Narrative Media Framing in Political Discourse (2506.00737v1)

Abstract: Narrative frames are a powerful way of conceptualizing and communicating complex, controversial ideas, however automated frame analysis to date has mostly overlooked this framing device. In this paper, we connect elements of narrativity with fundamental aspects of framing, and present a framework which formalizes and operationalizes such aspects. We annotate and release a data set of news articles in the climate change domain, analyze the dominance of narrative frame components across political leanings, and test LLMs in their ability to predict narrative frames and their components. Finally, we apply our framework in an unsupervised way to elicit components of narrative framing in a second domain, the COVID-19 crisis, where our predictions are congruent with prior theoretical work showing the generalizability of our approach.

Analysis of Narrative Media Framing in Political Discourse

The paper "Narrative Media Framing in Political Discourse" by Yulia Otmakhova and Lea Frermann provides a detailed exploration into the often overlooked field of narrative frames in automated media framing analysis. The research focuses on crafting a formal structure linking narrative elements to framing devices, which traditionally have been addressed more through topic-based methods.

Key Insights and Contributions

1. Framework Development: The authors introduced a sophisticated framework to analyze and predict narrative frames in media. This framework integrates key aspects of narratology with social and media studies on framing, facilitating the structural representation of narratives. The framework highlights particular roles such as Hero, Villain, and Victim, and how these are focused upon to convey moral evaluations, thus shaping the reader's perception.
2. Dataset Creation and Analysis: A dataset of news articles about climate change was annotated using the framework. The dataset was exploited to evaluate the prevalence of narrative frame components across different political leanings, illustrating how media outlets with various biases use narrative framing to promote specific interpretations.
3. Testing LLMs for Narrative Frame Detection: The paper rigorously tested the capability of contemporary LLMs to predict narrative frames and their components. This line of inquiry highlighted the challenges and potential improvements in automatic narrative prediction, particularly underscoring the models' limitations in accurately discerning nuanced narrative structures.
4. Cross-domain Application: Demonstrating the framework's versatility, the paper applied the narrative framing analysis to another domain — the COVID-19 crisis. It investigated politicians' speeches, revealing how different leaders employed narrative frames to communicate their strategies and evoke specific cultural stories during the pandemic.

Implications and Future Directions

The proposed framework carries significant implications for computational media framing analysis. It extends the scope of automated systems beyond superficial topic framing, offering deeper insights into how narratives shape political discourse. For practical applications, it could aid in detecting bias and understanding the underlying narratives in news articles or political speeches. Theoretically, the framework calls for refined approaches in NLP that consider narrative complexities, proposing that mere topic detection is insufficient for comprehensive framing analysis.

The limitations identified, particularly with LLMs not fully capturing narrative subtleties, suggest avenues for future exploration in enhancing these models. Introducing finer-grained narrative elements or hybrid approaches combining narratology with deep learning could improve frame detection performance.

Overall, this paper contributes to a deeper understanding of narrative in media framing and sets the stage for more nuanced automated analyses in political discourse, highlighting the dynamic intersection of narrative theory with NLP and the social sciences.