A shape grammar approach to computational creativity and procedural content generation in massively multiplayer online role playing games

Abstract

With rapid growth in both production costs and player populations over the last decade, the computer games industry is facing new scalability challenges in game design and content generation. The application of computers to these tasks – called procedural content generation – has the potential to reduce the time, cost and labor required to produce games. A range of generative algorithms have so far been proposed for procedural content generation. However, automated game design requires not only the ability to generate content, but also the ability to judge and ensure the novelty, quality and cultural value of generated content. This includes factors such as the surprise-value of generated content as well as the usefulness of content in the context of a particular game design. Studies of human designers have identified that the ability to generate artefacts that are novel, surprising, useful and valuable are facets of the human cognitive capacity for creativity. This suggests that computational models of creativity may be an important consideration for developing tools that can aid in or automate design processes. However, such cognitive models have not yet been widely considered for use in procedural content generation for games. This paper presents a framework for procedural content generation systems that use computational models of creativity as a part of the generative process. We demonstrate an example of such a system for generating instances for massively multiplayer, online role-playing games. The system combines the generative shape grammar formalism with a computational model of interest based on the Wundt curve to select new designs that are similar-yet-different to existing human designs. The approach aims to capture the usefulness and value of an existing human design while introducing novel or surprising variations through the model of interest. The system incorporates a metric that permits generated designs to be evaluated in terms of both their similarity to human designs and their novelty in the context of existing designs.

Introduction

Procedural content generation (PCG) is the programmatic generation or adjustment of the content of a computer game. Various aspects of PCG have been studied including online and offline PCG; PCG of different types of game content; and different classes of algorithms for PCG [1], [2]. One of the key challenges for all kinds of PCG lies in the need not only to generate game content, but to judge and ensure the novelty, quality and cultural value of generated content. This includes factors such as the surprise-value of generated content as well as the usefulness of content in the context of a particular game design. This paper proposes an approach to this issue by presenting a framework for PCG that generates new designs in a manner that can capture the usefulness and value of existing designs while also introducing novel or surprising variations. Specifically, we describe, demonstrate and evaluate an example of such a system for generating instances for massively multiplayer, online role-playing games (MMORPGs).

Studies of human designers have identified that the ability to generate artefacts that are novel, surprising, useful and valuable are facets of the human cognitive capacity for creativity [3]. Thus, to achieve these aspects in a PCG framework we draw on models of design cognition and creativity in humans and artificial systems. The approach in this paper provides a framework for building PCG tools that can take components of high quality designs produced by human designers and creatively recombine, reuse and augment those components using artificial creative processes to generate new designs. The approach aims to incorporate two broad aspects of creativity. First, usefulness and value are captured in a new design by ensuring that there is similarity or ‘closeness’ between new designs and one or more existing human designs. Secondly, novel and surprising variations are introduced in new designs using a computational model of interest to evaluate new designs.

Fig. 1 illustrates the high level framework for computational creativity in PCG proposed in this paper. First, to permit new designs to capture aspects of the usefulness and value of high quality human designs, our framework assumes the existence of at least one initial human design as a starting point for the creative process. This design is then decomposed (Step I) into primitive components that become the design variables input to the creative process (Step II). The creative process recombines those components to create new designs. This paper presents a tool that aids the human designer in Step I and automates the role of the human in Step II.

The remainder of this paper is organized as follows. Section 2 provides a cross-disciplinary review of literature in creative design and PCG. In Section 2.1 a number of generative algorithms and their role in PCG are briefly discussed. A detailed review of shape grammars is provided as background for the specific approach taken in this paper. As a basis for decomposing human designs into their constituent elements, Section 2.2 reviews existing taxonomies of the elements of game design. Section 2.3 reviews the elements of MMORPGs in particular, as the focus for the work in this paper.

Section 3 selects a subset of the elements of game content for MMORPG instances as the basis for a formal specification of our framework for computational creativity in PCG. Section 4 demonstrates an implementation of this framework and evaluates a set of designs for MMORPG instances it produces, including the design of the game space and the placement of non-player characters (NPCs) and interactive objects within the space. The paper concludes in Section 5 with a discussion of the implications and future directions of this work.