Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Cross-Domain Study of the Use of Persuasion Techniques in Online Disinformation

João A. Leite, Olesya Razuvayevskaya, Carolina Scarton, Kalina Bontcheva

TL;DR

The paper addresses how disinformation campaigns weaponize persuasion techniques across multiple domains, proposing a cross-domain framework that leverages a state-of-the-art persuasion technique classifier to analyze 16 techniques. It combines a multilingual RoBERTa-Large-based detection approach with four English-language disinformation datasets and a climate-change case study to reveal domain-specific usage patterns and contextual adaptations. Key findings include strong domain signals for techniques such as Repetition in Islamic issues, Authority in climate change, and Hypocrisy in the Russo-Ukrainian war, with a broader, domain-spanning presence of Doubt and Loaded Language. The work offers practical guidance for improving detection, counter-messaging, and media literacy, and contributes to reproducibility through open code and supplemental materials.

Abstract

Disinformation, irrespective of domain or language, aims to deceive or manipulate public opinion, typically through employing advanced persuasion techniques. Qualitative and quantitative research on the weaponisation of persuasion techniques in disinformation has been mostly topic-specific (e.g., COVID-19) with limited cross-domain studies, resulting in a lack of comprehensive understanding of these strategies. This study employs a state-of-the-art persuasion technique classifier to conduct a large-scale, multi-domain analysis of the role of 16 persuasion techniques in disinformation narratives. It shows how different persuasion techniques are employed disproportionately in different disinformation domains. We also include a detailed case study on climate change disinformation, highlighting how linguistic, psychological, and cultural factors shape the adaptation of persuasion strategies to fit unique thematic contexts.

A Cross-Domain Study of the Use of Persuasion Techniques in Online Disinformation

TL;DR

The paper addresses how disinformation campaigns weaponize persuasion techniques across multiple domains, proposing a cross-domain framework that leverages a state-of-the-art persuasion technique classifier to analyze 16 techniques. It combines a multilingual RoBERTa-Large-based detection approach with four English-language disinformation datasets and a climate-change case study to reveal domain-specific usage patterns and contextual adaptations. Key findings include strong domain signals for techniques such as Repetition in Islamic issues, Authority in climate change, and Hypocrisy in the Russo-Ukrainian war, with a broader, domain-spanning presence of Doubt and Loaded Language. The work offers practical guidance for improving detection, counter-messaging, and media literacy, and contributes to reproducibility through open code and supplemental materials.

Abstract

Disinformation, irrespective of domain or language, aims to deceive or manipulate public opinion, typically through employing advanced persuasion techniques. Qualitative and quantitative research on the weaponisation of persuasion techniques in disinformation has been mostly topic-specific (e.g., COVID-19) with limited cross-domain studies, resulting in a lack of comprehensive understanding of these strategies. This study employs a state-of-the-art persuasion technique classifier to conduct a large-scale, multi-domain analysis of the role of 16 persuasion techniques in disinformation narratives. It shows how different persuasion techniques are employed disproportionately in different disinformation domains. We also include a detailed case study on climate change disinformation, highlighting how linguistic, psychological, and cultural factors shape the adaptation of persuasion strategies to fit unique thematic contexts.

Paper Structure

This paper contains 8 sections, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Identifying Persuasion Techniques
  4. Disinformation Domains
  5. Analysis
  6. Domain-Specific Use of Persuasion Techniques (RQ1)
  7. Contextual Domain Adaptation (RQ2)
  8. Discussion & Conclusion

Figures (2)

  • Figure 1: Proportion of persuasion techniques on the different disinformation topics.
  • Figure 2: Top 10 most correlated LIWC features for each of the four domain-specific PTs in climate change compared to other domains. Statistically significant ($p{\mathrel{\mathord{<}}}0.05$) coefficients are indicated with an asterisk (*).