Laboratory for Memory and Digital Media Research
In close cooperation with the Institute of Physics of the Universidad de Antioquia (Medellín, Colombia), the lab will analyze social networks, memory research, and the general consumption of entertainment media. For this purpose, digital research methods (E-Research) will be applied to large amounts of data (Big Data) from social networks, alongside interviews, transcripts of selected telenovelas and news media, with the goal of rendering this data usable as a corpus of sources for the research project. The lab will apply automated Natural Language Processing (NLP) systems to memory and historical consciousness research, contributing to the development of an innovative methodology to identify, quantify, and analyze the following aspects within the digital corpus:
- Reception and processing of telenovela content by viewers,
- influence on the historical consciousness of the audience,
- dimensions, scope, and trends of trans-nationalization of memory images generated by telenovelas,
- effects on the political education of the audience.
To approach these four features in a first step, methods and algorithms of artificial intelligence will be used to develop cognitive agents that can capture and automate the analysis of large volumes of data. Cognitive agents are built on evidence-based reasoning (EBR), which is at the heart of problem-solving and decision-making tasks in a variety of fields, such as natural sciences, cybersecurity, archaeology, or peace studies. In this context, cognitive agents not only facilitate the quantitative evaluation of the source corpus, but furthermore, they complement their qualitative evaluation and visualize the behavioral dynamics of social networks. The methodological use of cognitive agents will continue to be increasingly important for the historical studies in the future due to the ever-growing amount of data, as the analysis of digital sources is no longer manageable with previous methods. This will also trigger the formulation of new research questions. The lab therefore also aims to develop new digital methods for historical research.
For this research project, four cognitive agents will be programmed. These will be guided by descriptors developed from interdisciplinary collaboration and research, as well as theoretical approaches to historical consciousness, memory studies, and civic education. The development of the descriptors will be based on the empirical materials already collected (from research on Chile, Colombia, and Brazil) as well as the data sets to be collected during the project. The descriptors will be explained more in depth in the following subsection on the methodological approach. Furthermore, approaches to reception research from communication studies and research contributions on political education in civil societies are used, which contribute to a better understanding of the influence of entertainment media on the political education of citizens.
Dörner, A. 2016. „Politserien: Unterhaltsame Blicke auf die Hinterbühnen der Politik“. Aus Politik und Zeitgeschichte. 66 (51): 4–11.
Koeman, V. 2019. “Tools for Developing Cognitive Agents”. https://doi.org/10.4233/uuid:f80750ee-db68-480e-8c58-2c167bd24ee5
Jensen, K. (comp.) 2012. A Handbook of Media and Communication Research. Qualitative and Quantitative Methodologies. London: Routledge.
Schulze Wessel, M. 2017. „Der Angriff des Populismus auf die Geschichte. Weshalb ein kritisches Geschichtsbewusstsein für die Demokratie unerlässlich ist“ Analysen & Argumente, Konrad Adenauer Stiftung, 256.
Tecuci, G.,Marcu, D., Mihai B., David S. 2019. “Toward a Computational Theory of Evidence-Based Reasoning for Instructable Cognitive Agents”. arXiv:1910.03990 [cs.AI]