Pedagogical application of game mediated stories

Much of what we currently know and understand about both education and storytelling is based upon the presupposition that are humans creating stories and other humans relaying that story and educating others about its context and meanings. Our most recent research explores what happens when the role of the human in creating those stories is minimised and / or deferred to a computer – this is artificial intelligence generating system-driven narrative, whereby procedural generation moves forward content via algorithms (see Travis, 2017). This procedural generation can produce consistent variables via ‘seeds’ which provide a uniform basis by which the procedural generation occurs which in turn produces the same outcomes (as per Minecraft), but that does to some extent act to limit the role that procedural generation (and thus the artificial intelligence) might have. Environmental storytelling is also a lot less overt and combined with procedural generation which is a partially randomised process it might be wise to consider procedural generation as the antithesis of any clear goal or attempt to create an objective understanding of the game mediated story in question.

On logistics and institutional responsibilities, multiplayer interaction is perhaps the least viable literacy practice to integrate into contemporary education systems, particularly those designed around students under the age of eighteen. Safeguarding is not something that can be flouted in the pursuit of a better understanding of game mediated stories. Therefore, while competition (as a form of challenge and story mediation) among students in games such as Among Us allows for engagement via competition and challenge as well as co-mediation of a unique story with each match the players engage in, the ability for players to interact with each other (and the need to) does present challenges to safeguarding and visibility that may be impossible to overcome. However, Among Us allows matches in closed lobbies via password-protection systems, which can facilitate small groups of students to work under supervision by a teacher in much the same way a group of students engaging in classroom play or a board game might be supervised. Generalising from this, we can surmise that system-driven narratives and non-creative computer systems such as procedural generation and AI response to players can be as predictable or unpredictable as an education system requires. If strict mediation and control of learner autonomy in the video game environment is applied through regular points of convergence and a potentially limited number of nodular situations are therefore applied in order to best suit our current practical implementation of education (particularly around that of English Literature), rule based story mediation within video games can produce easily predictable set outcomes that – while still mediating the story that is experienced by the player to an extent – will yield a story that educators and students alike can discuss without having watched or mirrored. This returns us to, and amplifies, the tensions around pedagogic ‘mastery’ discussed above with regard to L.A Noire.

If games might also hold educational potential over other media by using in-game challenges and allowing students to set their own goals and challenges without a teacher prompting them to follow a pre-set learning objective, then one might begin to question what role a rule and mechanic mediated story might hold in a contemporary learning environment rather than just a self-motivated recreational environment. Points of convergence lend linearity in story and as such, much clearer learning goals and objectives, while a mediated story – even if tightly controlled by rules and design mechanics – still offers a greater degree of interactivity and room for subversion of learning goals. Similarly, scoring systems and mechanics designed to influence player behaviour can be used to mediate said behaviour into pursuing the desired learning goals, but in turn, run the risk of becoming more of a focal point than the story itself. Perhaps, then, game rules and design mechanics might be used for engaging players (and learners) in something other than just the study of the co-mediated story that they work to create. Co-mediation assisted by a computer is not solely the domain of procedural generation however, and the level of intelligence that artificial intelligence exhibits in controlling ‘actors’ and the environment around it does mediate the course and outcome of the story.

Additionally, the ability of the AI to mimic human behaviour and create a fair but challenging opposition to the player helps to ensure the state of ‘flow’, whereby capability is matched by challenge to the extent that neither stress nor boredom result. This mirrors the objectives of good pedagogy, clearly. The state of flow which would similarly be sought after in a school environment requires more than just an adequate balancing of challenge to skill, but also the illusion of a tangible and possible story. If the system is expected to mediate a story via the likes of procedural generation and AI controlled actors, then the AI has to be able to act in a believable way in order to create and maintain flow in the story (and thus mediation of story) and to continue engaging the player in the story itself. If flow (see Owen, 2017:29) is one of the great gifts that game mediated stories might bring to education, then flow should be considered in terms of not only difficulty, but also in believability and student investment in the story that is being mediated and controlled by the system itself. In our static education system, the (human) author is taught alongside the context of their creation. The educator relies upon the text being the same for each and every student who they are teaching. Going beyond our previous work with L.A Noire as authorless literature, in the context of AI mediated stories, an author who lacks a distinct history and context due to being a computer and who only creates the story as the student plays presents an epistemological challenge in addition to the logistics already discussed, so in this case the acceptance of the hypothesis that games can be literature is also a greater leap.