This article has been published at the Dagstuhl Seminar 15051 "Artificial and Computational Intelligence in Games: Integration". The original publication, along with its bibtex entry and other information can be found here.

Procedural Content Generation and Formal Design
Joint work by:
Rafael Bidarra, Michael Cook, Antonios Liapis, Gillian Smith, Tommy Thompson, Mark Van Kreveld and Alex Zook

With the consolidation of Procedural Content Generation as an academic area of study, there now exists both a need for greater theoretical foundation for design and simultaneously an opportunity to build these theories. We posit that all procedural content generators are themselves encoding a formal theory of game design, in terms of both the process being followed and the products that are being generated. Understanding formal design theories (including their construction, analysis, and evaluation) requires integration of research and practice across several disciplines: the arts, humanities, design studies, and computer science. Game design and AI stand to benefit greatly from the varied perspectives on content generation by systematizing, generalizing, and deepening existing knowledge, while also broadening the range of topics addressed through procedural content generation.

Generators build upon design theories both explicitly and implicitly. Explicit models result from deliberate choices of a system designer to encode a form of the design knowledge for a domain (e.g., mazes) or a design process (e.g., the Mechanics, Dynamics, and Aesthetic framework). Implicit models are encoded within unacknowledged commitments expressed through algorithmic details (e.g., data structures, representation, generative methods) that may require critical analysis to be uncovered. Uncovering and acknowledging both the explicitly and implicitly encoded theories about the design process is key to learning from generators. Comparing them to each other and generalizing lessons learned from individual systems will lead to a broader, more inclusive, formal theory of game design. By examining the gap between generated products and exemplars made by human designers, it is possible to better understand the nature of the artifact being procedurally designed, thus building a formal theory of the artifacts being designed as well as the process taken to design them.

There is therefore value in understanding the design theory behind generators in terms of their goals, metaphors, methods, and ethics. Such a conscious commitment to an epistemological view on formal design theories in generators can lead to a better understanding of the generative process and the generated products. Formal approaches to design theories can support the analysis, interpretation and dissemination of PCG as an academic field and integrate practitioners including artists, designers, players, programmers, and researchers. For generative systems where the goal is primarily to produce a particular kind of playable experience, design theories allow artists to communicate their aesthetics, ethics, or message and reflect on it (in conjunction with the responses of a wider audience). For generative systems where the goal is primarily the system itself, design theories allow the creators of the system to fine-tune the algorithms as well as challenge current conventions.

In this working group we discussed motivations, high-level research topics, and initial research projects at the intersection of procedural content generation and formal design theory.

References

[1] Shaker, N. and Togelius, J. and Nelson, M. J. Procedural Content Generation in Games: A Textbook and an Overview of Current Research. Springer, 2015

[2] Togelius, J. and Yannakakis, G. N. and Stanley, K. O. and Browne, C. Search-based Procedural Content Generation: A Taxonomy and Survey. IEEE Transactions on Computational Intelligence and AI in Games (TCIAIG), volume 3 issue 3, pp. 172-186, 2011.

[3] Hunicke, R. and Leblanc, M. and Zubek, R. MDA: A formal approach to game design and game research. AAAI Press, 2004

This article has been published at the Dagstuhl Seminar 15051 "Artificial and Computational Intelligence in Games: Integration". The original publication, along with its bibtex entry and other information can be found here.