In the past decades, digital media have become an indispensable part of our contemporary societies. They have permeated myriad aspects of our social lives and they will keep on doing so in the future. Media, however, are non-neutral carriers of information. They structure communication according to their own logic. The current proposal asks how social media platforms are structuring communication when it comes to intergroup conflicts. More specifically, it proposes an in-depth study of the communication pertaining to the gamergate controversy; a controversy that erupted when a variety of people spoke out against the toxic and misogynistic aspects of gamer culture. Drawing on social identity theory and on the concept of network media logic, different kinds of threat-related communication will be linked to different kinds of social media characteristics. Methodologically, this project will combine analytical approaches such as machine learning, multilevel modeling and social network analysis. In the end, this research project will lead to theoretical and empirical insights in how opposing social groups communicate in mediated spaces. In doing so, an example can be set for future research looking into mediated intergroup conflicts. Considering the challenges of our society today (e.g., refugees, religious tensions) it goes without saying that this line of research will become increasingly important.