About

Yingdan Lu is an Assistant Professor in the Department of Communication Studies at Northwestern University. Her research focuses on digital technology, political communication, and authoritarian politics. She explores how authoritarian governments, such as China, utilize digital media and artificial intelligence strategically to maintain their rule, and investigates the downstream effects of these practices. Additionally, her work examines how individuals experience digital technology across different media environments and the impacts of these experiences. Yingdan employs both computational and qualitative methods in her research, including the development of new frameworks for analyzing multimodal data such as videos and smartphone screenshots, as well as extending natural language processing techniques to analyze cross-lingual digital communication. Her ongoing projects include pioneering methods for analyzing multimodal content to understand propaganda strategies in the digital age and addressing issues related to information manipulation and social inequality. Yingdan Lu holds a Ph.D. in Communication from Stanford University, where she also earned a minor in Political Science, along with an M.A. in East Asian Studies from Stanford and a B.A. in Journalism and Communication from Tsinghua University. Her research has been published in leading peer-reviewed journals and supported by notable funding sources, including the National Science Foundation and the Stanford Institute for Human-Centered Artificial Intelligence.

Research topics

• Political Science
• Law
• Business
• Sociology
• Social psychology
• Public relations
• Psychology
• Medicine
• Cognitive psychology
• Advertising
• Media studies

Selected publications

• An Open-Source Smartphone Screen-Recording Tool with Application to Political Science Research

  2026-04-23

  articleOpen access

  Political scientists seek to understand the political environment and to identify causal relationships within it. This often requires accurate measurement of how individuals consume political information. Many data-collection approaches, particularly those relying on social media content, infer consumption behavior from publicly posted material. Yet public-facing, platform-specific content misrepresents the fragmented, multimedia nature of smartphone use, and thus of political information consumption. Continuous smartphone screen-recording (CSSR) offers a theoretically strong means of capturing individuals’ information experience, but has remained largely inaccessible to political scientists due to technical, operational, analytical, and ethical barriers. We introduce an open-source CSSR tool to mitigate these challenges. We motivate its relevance for political science and outline its solutions to practical constraints that have limited CSSR’s adoption.

  PublisherDOI

• From Verification to Amplification: Auditing Reverse Image Search as Algorithmic Gatekeeping in Visual Misinformation Fact-checking

  arXiv (Cornell University) · 2026-03-10

  preprintOpen access

  As visual misinformation becomes increasingly prevalent, platform algorithms act as intermediaries that curate information for users' verification practices. Yet, it remains unclear how algorithmic gatekeeping tools, such as reverse image search (RIS), shape users' information exposure during fact-checking. This study systematically audits Google RIS by reversely searching newly identified misleading images over a 15-day window and analyzing 34,486 collected top-ranked search results. We find that Google RIS returns a substantial volume of irrelevant information and repeated misinformation, whereas debunking content constitutes less than 30% of search results. Debunking content faces visibility challenges in rankings amid repeated misinformation and irrelevant information. Our findings also indicate an inverted U-shaped curve of RIS results page quality over time, likely due to search engine "data voids" when visual falsehoods first appear. These findings contribute to scholarship of visual misinformation verification, and extend algorithmic gatekeeping research to the visual domain.

  PublisherDOI

• From Verification to Amplification: Auditing Reverse Image Search as Algorithmic Gatekeeping in Visual Misinformation Fact-checking

  arXiv (Cornell University) · 2026-03-10

  articleOpen access

  As visual misinformation becomes increasingly prevalent, platform algorithms act as intermediaries that curate information for users' verification practices. Yet, it remains unclear how algorithmic gatekeeping tools, such as reverse image search (RIS), shape users' information exposure during fact-checking. This study systematically audits Google RIS by reversely searching newly identified misleading images over a 15-day window and analyzing 34,486 collected top-ranked search results. We find that Google RIS returns a substantial volume of irrelevant information and repeated misinformation, whereas debunking content constitutes less than 30% of search results. Debunking content faces visibility challenges in rankings amid repeated misinformation and irrelevant information. Our findings also indicate an inverted U-shaped curve of RIS results page quality over time, likely due to search engine "data voids" when visual falsehoods first appear. These findings contribute to scholarship of visual misinformation verification, and extend algorithmic gatekeeping research to the visual domain.

  PublisherOA PDF

• An Open-Source Smartphone Screen-Recording Tool with Application to Political Science Research

  SocArXiv (OSF Preprints) · 2026-04-22

  preprintOpen access

  Political scientists seek to understand the political environment and to identify causal relationships within it. This often requires accurate measurement of how individuals consume political information. Many data-collection approaches, particularly those relying on social media content, infer consumption behavior from publicly posted material. Yet public-facing, platform-specific content misrepresents the fragmented, multimedia nature of smartphone use, and thus of political information consumption. Continuous smartphone screen-recording (CSSR) offers a theoretically strong means of capturing individuals’ information experience, but has remained largely inaccessible to political scientists due to technical, operational, analytical, and ethical barriers. We introduce an open-source CSSR tool to mitigate these challenges. We motivate its relevance for political science and outline its solutions to practical constraints that have limited CSSR’s adoption.

  Publisher

• Seeing the Surreal: Mapping Surrealism in Photorealistic AI-Generated Images Using Large Language Models

  Computational Communication Research · 2026-04-21

  articleOpen access

  Photorealistic AI-generated images (AIGIs) are increasingly indistinguishable from real photographs, raising significant social concerns. While prior research focuses on the production quality and detection of photorealistic AIGIs, such research often overlooks their expressive features. This study focuses on surrealism as a key feature of AIGIs, and introduces the concept of algorithmic surrealism to capture AIGIs' algorithmically driven and public accessible generative processes and consequences. Using 28,290 AIGIs collected from Instagram creators and a mixed-methods, Large Language Model (LLM)-assisted framework, we categorized physical, behavioral, and contextual surrealism at scale and found a pervasive presence of surrealism in AIGIs. Topic network and qualitative analyses show that algorithmic surrealism often appears in hybrid forms, indicates patterns of visual excess, reinforces stereotypes, transforms technical flaws into surreal aesthetic features, and exhibits visual homogenization tendencies. This study advances the theoretical understanding of surrealism and photorealism in the age of generative AI. Methodologically, it contributes to computational social science by demonstrating an LLM-based framework that integrates computational, qualitative, and network analyses to examine complex visual concepts.

  PublisherDOI

• Decentralized Propaganda in the Era of Digital Media: The Massive Presence of the Chinese State on Douyin

  SSRN Electronic Journal · 2025-01-01 · 2 citations

  articleOpen access1st authorCorresponding
  PublisherDOI

• Contextualizing Misinformation: A User-Centric Approach to Linguistic and Topical Patterns in News Consumption

  Proceedings of the ACM on Human-Computer Interaction · 2025-10-16 · 1 citations

  articleOpen access

  Exposure to misinformation poses significant challenges to democratic processes and public health, particularly during critical events like elections. This study adopts a user-centric approach to analyze the linguistic features of misinformation actually consumed by individuals during web browsing. Using data from a nationally representative panel of 1,240 American adults and their web-browsing data (21M URL visits) during the 2020 U.S. Presidential Election, we examine linguistic and topical differences in the content of 91K unique misinformation and hard news webpages by utilizing natural language processing techniques and Large Language Models. We find that misinformation consumed by users is generally easier to read, exhibits higher negative sentiment, and employs more moral language than hard news. We also find significant linguistic variations across topics--misinformation can be diverse and vary in linguistic features depending on the subject matter. We also identify heterogeneity across key user characteristics: older adults consume more misinformation about COVID-19 and health, with content showing more negative sentiment and fewer moral terms than expected. Republicans engage with misinformation characterized by more negative sentiment and higher moral language, focusing less on health topics and more on social and political issues. These results highlight the importance of a user-centric approach and suggest that interventions to combat misinformation should be tailored to specific topics and user characteristics for greater effectiveness.

  PublisherDOI

• Baseline and changes in prognostic nutritional index associate with heart failure hospitalization and all-cause death in patients with cardiac pacemaker

  BMC Cardiovascular Disorders · 2025-03-31 · 3 citations

  articleOpen access

  OBJECT: The study evaluated prognostic nutritional index (PNI) levels and changes as predictors of heart failure hospitalization (HFH) and mortality in cardiac pacemaker patients. METHODS: PNI was calculated on admission and at the end of the 1-month follow-up, and their net changes (Δ) were calculated by PNI at follow-up minus the corresponding PNI on admission. The optimal cutoff value of baseline PNI was determined by the ROC curve. Based on this optimal cutoff value or the change in PNI during follow-up, patients were divided into high-PNI and low-PNI groups, or into improved PNI group (ΔPNI > 0) and deteriorated group (ΔPNI ≤ 0). The crude and adjusted cox proportional hazard models were used to analyze the associations between adverse events and PNI or ΔPNI. Restricted cubic splines were used to model the association of the PNI-endpoint events. RESULTS: A total of 927 patients were enrolled in the study. The risk of HFH and all-cause death remained significantly higher in patients with low PNI, even after adjusting for other risk factors (hazard ratio [HR]: 1.997, 95% confidence interval [CI]: 1.220 - 3.203, P = 0.006; HR: 2.501, 95% CI: 1.392 - 4.494, P = 0.002, respectively). Patients with ΔPNI ≤ 0 faced higher risks of HFH and mortality compared to those with ΔPNI > 0(HR: 3.146, 95% CI: 2.024 - 4.892, P < 0.001, HR: 2.082, 95% CI: 1.223 - 3.544; P = 0.007, respectively) CONCLUSION: Low baseline PNI and ΔPNI ≤ 0 during follow-up effectively predicted HFH and mortality in cardiac pacemaker patients. CLINICAL TRIAL NUMBER: Not applicable.

  PublisherOA PDFDOI

• Crafting Synthetic Realities: Examining Visual Realism and Misinformation Potential of Photorealistic AI-Generated Images

  2025-04-25 · 5 citations

  articleOpen access
  PublisherOA PDFDOI

• Unmasking Deception

  2025-07-15

  book-chapter

  Journalists in the digital media era have faced many challenges – including sorting online misinformation from accurate information in news reporting. In particular, the increasing presence of visual misinformation can be detrimental to journalism as visual information is especially powerful to audiences, and penetration of visual misinformation could potentially lead to significant distortions in audience perception and understanding of the issues journalists report. Various types of computer vision techniques have been developed to mitigate the influence of visual misinformation circulated online. This chapter explores computational models designed to detect visual misinformation and discusses how these computer vision tools can empower journalists in their efforts. It starts by explaining why it is important to understand and detect visual misinformation in the digital age. It then systematically reviews different types of online visual misinformation and their impact on journalism. The chapter further examines various computer vision techniques that automatically identify visual misinformation, highlighting their potential to assist journalists in identifying visual misinformation. Despite their merits, it concludes by identifying the limitations of current computer vision tools and stresses the importance of visual literacy education for journalists and audiences to combat online visual misinformation.

  PublisherDOI

Frequent coauthors

• Jennifer Pan

  27 shared

• Yilang Peng

  University of Georgia

  6 shared

• Cuihua Shen

  University of California, Davis

  3 shared

• Leo Yeykelis

  3 shared

• Yiqing Xu

  Stanford University

  3 shared

• Jungseock Joo

  Nvidia (United States)

  2 shared

• Kunwoo Park

  Institute for Basic Science

  2 shared

• Daniel Muise

  2 shared

Education

• PhD, Communication

  Stanford University

  2023

• Master of Arts, Center for East Asian Studies

  Stanford University

  2017