Don't just watch, join in: Exploring information behavior and copresence on Twitch

This mixed-methods study examined users' information behavior and their perceptions of copresence on Twitch. tv, where millions of people come together live every day to stream, interact, and make their own entertainment. Human information theory model and social identity theory constituted the theoretical framework for this research. Studying topic specific live streaming services is an emerging and exciting field in communication. Topic specific live streaming sites such as Twitch.tv are evolving constantly into important sources of information that complement the traditional information systems such as libraries and online search engine sites like Google. Chat logs of three live streams on Twitch.tv were analyzed using Linguistic Query and Word Count (LIWC) and SPSS statistical tools. Qualitative thematic analysis was carried out using Nvivo 12. Information reaction and production were the most frequent information behaviors across three streams. Qualitative analysis indicated that users within the three live streams showed a great deal of copresence. This study is an important first step to provide theoretical insights into understanding human information behavior on Twitch, topic specific live streaming sites, and social live streaming sites in general.