Complexity, Mystagogy, and the Limits of the 'Analytical' Learning Cycle

This paper, by Feng Zhu of King's College London's Department of Digital Humanities, is a theoretical argument about how players learn computer games and what that learning means politically and aesthetically. It challenges the view, associated with Fredric Jameson's notion of "cognitive mapping," that games can cultivate a comprehensive, top-down understanding of complex systems that supports socialist political consciousness. Zhu's central thesis is that the learning cycle in games reveals both cognitive and non-cognitive modes of comprehension, and that the non-cognitive side, which Bernard Stiegler terms "mystagogy," is the more significant aesthetic and political contribution of games.

The argument rests on several supporting observations. Drawing on Juul, Salen and Zimmerman, and Arsenault and Perron, the paper describes a shift from bottom-up analysis of individual elements toward top-down reliance on a mental model of the game system as players gain skill. Zhu then cites computational studies by Aloupis, Demaine, and Guo (2012) and Viglietta (2014) establishing that classic Nintendo and 1980s–1990s games are NP-hard, meaning optimal play cannot in principle be determined in finite time. From this, the paper concludes that a complete cognitive map of such games is unattainable. The continuing discovery of glitches in Super Mario Bros. by speedrunning communities, documented some thirty-three years after its release, serves as an illustrative case.

The paper's scope is theoretical and interdisciplinary rather than empirical. It engages film, media, and critical theory alongside game studies and computational complexity, and it does not present survey data, sample sizes, or original experimental findings. Its methodology is conceptual analysis of existing literature, with case examples drawn from specific titles and community practices.

The paper's conclusion is that the political potential of games lies not in enabling cognitive mapping but in cultivating reflective awareness of how we make sense of systems we cannot fully comprehend. It also suggests that the analytical logic embedded in games, which aligns with informatic and control logics, may be a target for resistance rather than a model to be emulated.