Charting Performative Possibilities in Games: A Large-scale Census of Characters and Representation

Watch Session:



It is anecdotally understood that games currently provide a limited set of representative and performative possibilities to players: who is the avatar you control? Are they white? Black? Hispanic? Male? Female? Queer? Straight? Old? Young? The characters we play, as encoded within the formal system of the game, often do not look like us or have backgrounds we can relate to. In this project, we seek to move beyond anecdotal accounts of limited representational possibilities in games to develop a census of playable characters in games that allows us to quantify the formal demographic possibilities afforded by the art form. This kind of census work actually requires three different, interrelated, studies, as there are three primary ways that the characters we control in games are designed:

  • Single Default Characters – The player is given no choice of who their character or avatar will be.
  • Multiple Default Characters – The player selects between a set of pre-defined characters.  These characters may differ only cosmetically, or they may have different game functions, or narrative roles, within the game world.
  • Parametric Characters – The player configures many aspects of the character’s appearance. Parametric characters may also be configurable in ludic and narrative ways that complicate representation: character classes often come with representational baggage independent of other character configuration choices.

A census of this type has not yet been made of playable character possibilities, especially at our scale, at currently 200 games. “The Virtual Census” analyzed playable and non-playable characters during the first 30 minutes of 133 games released during 2005 to determine “gender, race, and age in comparison to the US population” (Williams et. al. 2009). The LGBTQ Video Game Archive (Shaw et. al. 2016) is also concerned with any queer characters within games. Our study is specifically concerned with who players can inhabit in-game, rather than who they may interact with. While representation is only one factor for considering how games can affect players, the results of this study should provide a landscape of possible representations, within which discussions of identification, affect, individual experience, and the systematic marginalization of certain representations or identities can occur.



Dan Gardner is an Informatics PhD student at UCI. He is Dan Gardnera member of several labs at UCI including EVOKE, TECHDEC and the Transformative Play Lab. His research interests are in how we interact with digital media, rather than through it, and how media design itself can exert authority over users. He is interested in how scholars leverage the affordances of digital technologies in order to develop new methods of knowledge collection, creation, and narration, particularly through collaborative means. Dan brings experience from his professional history of digital animation, video game retail, and Information Technology (IT) security analysis to bear in his work.


Dr. Joshua Tanenbaum is an Assistant Professor in the Department of Informatics at UC Irvine. He is a member of the UCI Institute for Virtual Environments and Computer Games, the Laboratory for Ubiquitous Computing and Interaction, the EVOKE Lab, and a founding member of the Transformative Play Lab. His research includes studies of agency and identity transformation in games and digital narratives, maker and DIY subcultures, design fictions and future oriented human computer interaction, and tangible, wearable, and ubiquitous computing. In his (theoretical) spare time he creates steampunk artwork and costumes, makes games and occasionally writes music.