Enriching the Metapicture: How to Analyse Audio-Visual Content on TikTok and Other Platforms

The paper discusses a methodological approach to the study of audio-visual content on platforms like TikTok. We argue that existing methodologies for the analysis of visual content based on the notions of the metapicture are not yet fully equipped to account for the multi-layered, dynamic, and information-rich nature of audio-visual material. To fill this gap, we propose an approach aimed at ‘enriching’ the metapicture with textual and other components, alongside visual ones, to critically and systematically analyse video collections. First, we explain how researchers can enrich the metapicture by following three main principles: metadating, disaggregation and reaggregation, and readability. To illustrate our approach, we present a case study based on data donations to explore the content and gender representations TikTok users are exposed to. The study is structured in three steps, each introducing a type of enriched metapicture: the megapicture, scene-by-scene timeline, and video stack.