A MMORPG Prototype for Investigating Adaptive Quest Narratives and Player Behavior Emmett Tomai University of Texas – Pan American 1201 W. University Dr. Edinburg, TX 78539, USA tomaie@utpa.edu Rosendo Salazar University of Texas – Pan American 1201 W. University Dr. Edinburg, TX 78539, USA rsalazar6@broncs.utpa.edu Dave Salinas University of Texas – Pan American 1201 W. University Dr. Edinburg, TX 78539, USA dsalinasez@broncs.utpa.edu ABSTRACT In this paper, we present a research agenda towards adaptive quest narratives in MMORPG shared worlds. There is a significant incompatibility in these games between the dominant quest narrative model that guides and motivates gameplay, and player driven change in the world. We discuss incremental extensions to improve the trade-offs in that conflict, and present a simulation system for principled exploration of aggregate player dynamics. This situates and motivates the need for player behavior models and collection of player decision data. We discuss the development of an interactive prototype for gathering this data and evaluating novel extensions to MMORPG game mechanics. Categories and Subject Descriptors J [Computer Applications] General Terms Algorithms, Measurement, Design, Human Factors. Keywords Games, interactive narrative, simulation. 1. INTRODUCTION The MMORPG genre has seen dramatic growth in the past decade. But while many diverse designs exist, the explosive popularity of first Sony’s EverQuest and then Blizzard’s World of Warcraft has entrenched a dominant model. One of the core elements of this model is quest-driven gameplay. In brief, players control avatar characters in a shared, persistent world. They are free to roam in the world, interacting with the environment and other players as they see fit. Throughout the world there are system-controlled entities (Non-Player Characters or NPCs) that the player can interact (speak) with to receive quests. These quests specify task requirements (e.g. kill 10 rats) and rewards (e.g. progress points and a shiny hat), and provide narrative text intended to situate and motivate the task (e.g. help us, the rats are eating all our food!). Multiple quests can be accepted, worked on, put aside, abandoned or completed at the players’ discretion. This non-linear, on-demand system of guidance and motivation has proven remarkably effective, in spite of text that is often trite and formulaic and tasks that are highly repetitive. This does not mean that game designers or audiences are content with the quest model. It has significant limitations, one of the most striking of which is that it forces the shared world to be mostly unchanging, in order to support multiple players working on quests at different times. Interestingly, the most successful games have lived with this limitation, making static worlds. But developers continue to look for creative ways to add dynamic change in response to player actions. Two of the biggest MMORPGs, BioWare’s Star Wars: The Old Republic and the later updates to World of Warcraft, have heavily used techniques to separate players into copies of the world at key times, enabling localized, non-shared change. In contrast, Trion Worlds’ Rift, one of 2011’s significant successes, attempted to make a more dynamic world, where player actions freely change the world. They omitted quest text to make this work, but found in testing that players felt disconnected from the world. They re-added quests, creating two layers of gameplay: one static and one dynamic, with limitations on the latter [14]. For MMORPGs, it appears that the ability to tell motivating stories is highly valued and necessary, yet there is clear initiative to provide more dynamic, responsive worlds for an increasingly experienced audience. The established quest model provides a simplistic but proven way to adapt static narratives to players’ schedules and interactions in a mostly static shared world. We propose an agenda of incrementally increasing both the dynamism of the world (in response to player actions) and the ability of quest narratives to adapt accordingly. In this paper, we present supporting work for this agenda. Extending the quest model is not in itself difficult, but those changes must ultimately be evaluated according to their impact on player experience. And every player’s experience is impacted by aggregate player dynamics that can be very complex and non-obvious. Evaluating these extensions in a live game would be ideal, but is highly impractical. Even focused, small-scale controlled testing would require a significant number of players and amount of time. To address these challenges, we have created a lightweight MMORPG simulation for predicting aggregate player dynamics. This simulation is being integrated into a prototype system for small-scale play testing, which will be used to collect player behavior data. This data is necessary to generate player models and investigate whether we can accurately simulate aggregate player decisions. We describe these combined systems in the context of a simple extension to the standard quest model. 2. RELATED WORK There has been a rise in research on MMORPGs as the genre has grown, primarily focused on broad player motivations and behaviors (cf. [13]), and correlating them with practical, external concerns like network load [2], addictive behavior [4] and real- Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. FDG’12, May, 2012, Raleigh, NC, USA. Copyright 2012 ACM 1-58113-000-0/00/0010…$10.00.