Selling and Buying Visual Media Frames
Speaker: (opens in a new window)Olga Gasparyan (Hertie School)
Wednesday, September 21, 14:00–14:45 (Irish time)
Please register (opens in a new window)here to receive the link and password to the online meeting and information on the room at UCD.
Abstract: Numerous studies have examined media bias mostly paying attention to how news outlets manipulate information to present a desired picture of the world. However, selling a particular picture makes the most sense if people buy it, i.e. adjust their perspectives according to the objectives and desired direction of the seller. Thus, when exploring media frames, it is crucial to not only pay attention to how the media frames are being sold (whether and how outlets introduce particular frames) but also to a buyer's perspective (whether and under which conditions people interact with said frames). In this paper, we focus on both perspectives. First, we explore whether media outlets with different political views use systematically different visual representations of a polarizing issue, immigration, using a series of computational tools for image analysis. In the second phase, we conduct a survey to determine if and how people decode the visual patterns of outlets with distinct ideological slants. We find that media outlets selectively amplify visual representations that are more likely to activate partisan stereotypes. However, although we find evidence that visual content trigger ``polarizing'' reactions between partisans even when the source of this content is not revealed, we also find several instances when this does not occur. With this study, we contribute to the literature that examines the relationship between the use of visuals and ideological factors. In future work, we will explore the effect that visual content have on attitude formation, and the interplay of this element with text.
About the speaker: Olga Gasparyan is a Postdoctoral Research Fellow at the Data Science Lab of the Hertie School and an Academic Coordinator of the Data and Methodology Center at the SCRIPTS Cluster of Excellence. She received her PhD in Political Science from the University of Rochester. Her research lies in the fields of comparative political economy, public policy, and media studies, and is conducted with an application of causal inference and machine learning and deep learning models. She is a co-organizer and a coordinator of the Hertie Data Science Summer School, a member of the CIVICA Data Science Seminar Series org-team, and a co-investigator on the 2022 Horizon Europe Grant with the ''Climate Action To Advance Healthy Societies in Europe” (CATALYSE).