The Yale School of Management presents
Michelle Torres, Assistant Professor, Department of Political Science, University of California, Los Angeles:
“Beyond Prediction: Identifying and Accounting for Latent Treatments in Images.”
Images are a rich and crucial element of political communication. The complexity of the information they convey creates challenges for the identification, interpretation, and explanation of the effects of visual messages on information processing and attitude formation. In this article, we adapt a methodological approach used in text analysis, the supervised Indian Buffet Process (sIBP) developed by Fong and Grimmer (F&G, 2016, 2021), to identify latent treatments in images and evaluate their impact on outcomes of interest. First, we use a convolutional neural network (CNN) to decompose images into substantively meaningful and interpretable tokens, visual words, to then form the input of the sIBP. Then, we follow the framework introduced by F\&G and demonstrate the utility of this approach using two datasets: 1) a novel experiment measuring attitudes towards climate change in response to visual frames and 2) images of the Black Lives Matter (BLM) movement protests manually labeled by human coders according to the level of conflict they depict. We find significant differences between demographic and political groups in the way they perceive images, and also unmask latent treatments that confound the relationship between our treatment and outcomes of interest. Importantly, this paper extends the usage of computer vision tools in social sciences beyond prediction of image labels to uncovering, understanding, and visualizing the features of images that produce outcomes.
Michelle Torres holds a Ph.D. in Political Science and an A.M. in Statistics from Washington University in St. Louis. Her broad research interests are in the fields of political methodology and political behavior, with a special emphasis on computer vision, causal inference, public opinion, and political communication. She is interested in making statistical and computer science methods accessible to political science, and in developing and applying innovative and rigorous tools to achieve a better understanding of social issues, especially in the fields of political behavior. Her current agenda focuses on the analysis of images and pictures using computer vision and machine learning techniques in order to classify political visual messages/frames and understand their role in the generation and processing of political information. She also works on the intersection of causal inference and machine learning.