The Perception of Roles In Game Systems

Introduction

In “Communication Design”, professor Jorge Frascara [2007] states that all design generates communication situations – not only objects, but events in which people interact with the design. The idea of the designer as a communicator isn’t novel in any way: the act of design is, by morphology, associated with designate – to signify, to imbue with meaning. Nonetheless, it’s a concept rarely discussed in the field of game design.

Frascara’s work has another concept that is even more interesting to game designers: the most important aspect of the communication between the designer and the public is not the act of communication itself, but the impact that it has “on the knowledge, the attitudes and the behavior of people” [Frascara 2007]. In this article I propose a few lenses to look at how the different ways a game designer can communicate his or hers systems may affect players’ mental models and, consequently, the way that the game experience plays out.

In the next sections I’ll try to outline the concept of roles of objects in a game system, the basics of space of possibility and decision-making in games, and how the perception of roles change the game experience. I will also do case studies of successful games such as Relic’s Dawn of War 2 and Richard Garfield’s Magic: The Gathering.

Semiosis and Choice

All human beings live in worlds they construct for themselves by semiosis, defined by Peirce as the production and interpretation of signs [Atkin 2010]. Perception is our only interface with the “real” world around us, and while the actual process of semiosis is infinitely complex and sometimes chaotic, the underlying concept is a fairly simple triadic representation: a sign (object) mediates what it represents (or what it is intended to represent) and the way it is interpreted (the impact on the interpretant).

Fig. 1 Adapted from Lucy Niemeyer’s presentation “Semiótica Aplicada ao Design”

The act of playing a game, on the other hand, is composed of a series of decisions that are taken based on, roughly, four steps that players take:

1)    Interpretation of the current state of the game;
2)    Interpretation of the space of possibility of the current state;
3)    Model of an ideal future state of the game;
4)    Judgment of the most adequate action or set of actions to achieve the desirable state;

While it is common to believe that playing a game “well” means excelling at the last step of decision-making, it is interpretation that provides the basis to work with. That is, a wrong interpretation of the game state and/or of the space of possibility may lead the entire process towards frustration and “bad decisions”. The first two steps, then, are more interesting for our discussion, as they form players’ mental model [Norman 1988] of the game. A mental model is a “thought reconstruction” of how something works in the real world, built by experience, training and instruction [Norman 1988] – a model built by semiosis.

It is pertinent, then, to discuss game state and space of possibility a little more in-depth. The game state is clearly and concisely explained in “Rules of Play”:

“(…) the game state represents the current condition of the game at any given moment. In a Chess game, for example, the game state is represented by the arrangements of the pieces on the board, the captured pieces, and which player is about to move next. In a console fighting game such as Virtua Fighter 4, the game state includes which two combatants were chosen, the health and other fixed and variable stats of the two fighters, their relative spatial positions, and the arena in which they are fighting. The game state is a formal way of understanding the current status of the game, and does not take into account the skills, emotions, and experience of the players. (…) the game state refers only to the formal, internal condition of the game.”

[Zimmerman and Salen 2004]

Space of possibility is the space of possible action that players explore in a specific moment in the game [Zimmerman and Salen 2004]. In a game of chess, for example, it covers every legal movement of every piece that the player controls within a game state. It is the collection of all possible actions and outcomes inside the designed space of the game – all actions and outcomes artificially made possible by the system [Venturelli 2009]. Interpretation of the space of possibility is a concept deeply related with the idea of affordance -  everything that an object or system can offer, support, furnish [Gibson 1986].

Roles In Game Systems

Every game is a system, thus every object that composes it has a role to play on that system. A role, as I’ll develop throughout this article, can be more or less subjective, more or less dynamic, but it represents a meaningful way with which an object affords change or prevention of change on the game state. In badly designed games (or games that have been deliberately designed like this), not all objects have roles. Ideally, good game design guarantees that all objects are integrated into the system to provide players with meaningful decisions [Zimmerman and Salen 2004], that is, decisions which affect the game state in a way that influences future decisions.

Roles can be designed to be more or less extrinsic or intrinsic. It is simpler to understand the concept by analyzing the extremes. Completely intrinsic designs creates constructed, explicit, static game experiences. Completely extrinsic designs create emergent, implicit, dynamic experiences.

The classic game of Rock, Paper, Scissors is an example of a completely intrinsic role design. There are three objects in the game system, and their roles are as follows:

1)    Rock: beats Scissors, gets beaten by Paper, ties with Rock;
2)    Paper: beats Rock, gets beaten by Scissors, ties with Paper;
3)    Scissors: beats Paper, gets beaten by Rock, ties with Scissors;

Each object has one clear role that players must know to understand the rules of the game, these roles never change, and objects never have different roles. This game is, then, a completely intrinsic role design – all roles are determined by the system regardless of player strategies, mental models and game state.

It is important to note that objects in the system are not the only element that affects the way decisions are made. For example, in the game of Rock, Paper, Scissors there is a very simple layer of metagame that makes it fun – thus, playing Rock just after a Rock vs. Rock play has its own meaning regardless of role. Just as it would if you were playing against an opponent that is known for always starting the game by playing Rock. The concept of roles is powerful and useful, but it only exists on the system level of a game.

On the extremely extrinsic role designs side, the ancient game of Go is one of the best examples. In Go, every object is symmetrical to each other – there is no “special” piece. Roles exist, but they are completely determined by player strategy and game state. For example, a piece put near the end of a clear diagonal has a role, on that game state, of “marking the territory”. This role is completely situational, since another piece put next to it may invalidate or change it completely. The system does not dictate roles out of the box: they happen dynamically during gameplay.

Roles are not to be mistaken by mere asymmetry between objects, but asymmetry does play a part in making role designs less extrinsic. In Chess, for example, each type of piece has its own rules of movement, but theoretically any piece can be used to capture any other and to execute any play. In practice, however, the asymmetry of movement and starting positions of the pieces creates unique roles for them in gameplay, even though many overlapping roles exist.

Most designs sit in the middle of the two extremes. Classic RPG combat, with its “Tank-Healer-DPS” roles, is more intrinsic, while Chess or strategy games such as Starcraft are more extrinsic. The concept of dynamic extrinsic roles will be further developed in the next section.

The Perception of Roles

Playing a game is a dance of pattern recognition – it’s about signifying and creating mental models. Our brain is pretty good at that: “the destiny of games is to become boring, not to be fun. Those of us who want games to be fun are fighting a losing battle against the human brain because fun is a process and routine is its destination (...). All of this happens because the human mind is goal driven” [Koster 2005]. As I’ve wrote previously:

“Our brains are constantly trying to optimize information, simplify it, put everything in little boxes inside of our heads. We create and refine patterns throughout our lives to identify facial expressions, written and spoken words, odors, everything that can be perceived by our senses. It is through this process that we have meaning. Playing games is not different. When playing Space Invaders, for example, a player is constantly trying to figure out the patterns for the enemy ships' movements, the trajectory of his shots, the best way to destroy the first rows of enemies as fast as possible, and so on.”

[Venturelli 2009]

The artificiality of game systems is one of the main reasons why this intense process of semiosis occurs – and arguably why the act of playing games is so compelling. Zimmerman and Salen [2004] relate the process of entering a game as a step into a “magic circle” where objects have new meaning – a pair of sandals is no longer a piece of attire when it is used to mark the scoring zone of a Sunday soccer game.

“What does it mean to enter the system of a game? How is it that play begins and ends? What makes up the boundary of a game? (…) At stake is an understanding of the artificiality of games, the way that they create their own time and space separate from ordinary life. The idea that the conflict in games is an artificial conflict is part of our very definition of games.”

[Zimmerman and Salen 2004]

And if a game system is its own universe, when in contact with a game for the first time we act not unlike the whale from the Hitchhiker’s Guide, investigating our new surroundings, wondering what everything is and what it is for, and giving each object a sign that seems in line with what it affords us.

“Why am I here? What’s my purpose in life? What do I mean by who am I? Calm down, get a grip now … oh! this is an interesting sensation, what is it? It’s a sort of … yawning, tingling sensation in my … my … well I suppose I’d better start finding names for things if I want to make any headway in what for the sake of what I shall call an argument I shall call the world, so let’s call it my stomach. (…) And wow! Hey! What’s this thing suddenly coming towards me very fast? Very very fast. So big and flat and round, it needs a big wide sounding name like … ow … ound ... round ...  ground! That’s it! That’s a good name – ground! “

[Adams 2002]

If roles, as discussed earlier, represent ways that objects afford change or prevention of change on the game state, they are a very powerful way to understand how pattern recognition causes refined mental models to emerge when we play. According to Koster [2005], “the natural instinct of a game player is to make the game more predictable because then they are more likely to win”. This means we are constantly optimizing our thought processes to include as much useful information as possible in as little signs as possible in order to predict accurately how each decision can affect the game state. That’s how a player grows increasingly skilled in a game, replacing the raw presentation of the design with a highly efficient model which only takes into consideration the elements that meaningfully affect decision-making.

When playing a game for the first time, our repertoire of objects and their roles is fairly big – even if we are yet to acknowledge some of them (“what is this box for? What about that pipe?”). We are not sure of how each object interacts with one another, what is important, what is not important, what is possible and what is not possible. With experience, training and instruction we slowly start to refine our model of the game: different objects start to be grouped under the same category if they all affect the game in pretty much the same way (“these are all objects that I can use to climb”); new patterns that previously were not even perceived now take most of our attention as to when they will appear and how (“maybe that noise I heard means that a climb-able object appeared over there”); common situations are more easily identified, and roles are generated to resolve each one (“I need a climb-able object to go over there”). 

In more “challenging” games, identifying roles is more difficult. Above, when describing the design of roles in Go, I stated that the same piece in the same position in the same game can participate in several different “plays” at the same time, and have different roles in each of them. One particular object in a system may have multiple – even simultaneous – roles (“I can also use that pipe to divert water”), or change roles according to the game state (“If I hit that platform, I will be able to climb on it”).

Case Study: Dawn of War 2

Dawn of War 2 is a real-time strategy game developed by Relic Entertainment. Players form small armies based on the classic miniatures game “Warhammer 40K” and, in the title’s main competitive mode, battle for control of strategic points. Dawn of War 2 is a very interesting example of role design: there are several asymmetrical armies with asymmetrical units at players’ disposal, all with many upgrade options that unlock different capabilities as each match progresses.

Below the surface, however, a very clear and simple set of roles governs most in-game decisions. When players, for example, see a squad of units called “Slugga Boyz” charging against an opposing squad of “Devastator Heavy Bolter” troops, they can already be sure that the latter will win. What they are really seeing is not necessarily a network of relationships between each individual unit, but instead a simple case of “melee” going up against “suppression”.

 “Melee” and “Supression” are both examples of the main combat roles that units play in the game of Dawn of War 2, which also include “Jumpers/Teleporters”, “Vehicles”, and others. Roles are dynamic, meaning that the same unit can have multiple roles, or even change roles depending on the situation and/or in choice of upgrades by the player. This type of quasi-intrinsic role design is of great use for designers to proper balance the otherwise extremely asymmetric armies and units in the game, but it also eases the learning curve that normally keeps less dedicated players away from other titles on the same genre, such as the classic Starcraft, developed by Blizzard.

Interestingly, the recognition of this fact led Relic’s designers to actually try and communicate the roles in their game directly to players. A few months after Dawn of War 2 was released, an update became available which would show small icons called “Decorators” on top of each unit representing their respective main roles.

Fig.2 Dawn of War 2 Decorators

This decision had little impact on high-level play – skilled players already understood the game way better than the icon set could ever present – but made the learning curve even lighter for new users since it was no longer required of them to carefully examine each unit in the game before making their own decisions about their roles. A quick glance was enough for a rough assessment. It also lead them easily towards the mental model that the designers intended for the game.

Of course there is a lot more depth to the actual system than this – the dynamic nature of roles in the game was already mentioned, and combat is only one of the title`s strategic elements. Nevertheless, Dawn of War 2 is one of the finest examples of how role design and communication can deeply affect player experience.

Case Study: Magic: The Gathering

Magic: The Gathering is a collectable card game designed by Richard Garfield and published by Wizards of the Coast. In it, players assemble decks of around 60 cards from a collection of thousands and match them against other decks to test not only their playing skills, but their “deck-building” skills as well.

Magic is a good example for this discussion, as the gargantuan space of possibility when building a deck can only be resolved without frustration by players with a solid role-based mental model. It is easy, however, to be confused by the game`s own categorization of cards. Cards in Magic have “types”, which dictate unique rules for those groups of cards. These types can be “enchantments”, “instant magics”, “terrains” and so on. Types, however, are not directly related to roles.

Magic has no formal way of communicating roles to players – they have to find them on their own. That`s one of the main reasons why the game is considered to be extremely reliant on metagame knowledge to function – modern gameplay of Magic is actually designed to work like this, and players and designers alike take pride of that characteristic, even if the tough learning curve and seemingly imbalanced state of the game may look roughly designed by an outside observer.

Community, in these cases, is always helpful in guiding players towards the “ideal” mental models of the game. There is a very interesting mode of playing Magic called “Draft”, in which players received sealed packs of random cards and must all take turns picking cards from this pile to build their own decks, taking into consideration not only an efficient fulfillment of all the roles they`ll need in gameplay, but also trying to deny the most useful cards to their future opponents.

Fig.3 The card is described as a “red instant magic”, but its role is actually “Removal”

In such a seemingly chaotic semantic environment as “Draft”, roles play a very important part. The community even have clear names and references to some of the most important ones. For example, cards which roles are to remove other dangerous cards from play are called “Removals”; cards which allow players to overcome other players` defenses are called “Evasion”; very rare cards which can win the game if brought into it are called “Bombs”, and so forth and so on. This is paramount for players as they build their decks, as they will continuously assess their mental models in ways such as “do I have enough Removal on my deck?”, and “I think I need some ways to provide me Card Advantage”. It is a much simplified and manageable model to deal with a huge amount of sometimes unknown cards than it would be if players had to judge the individual value of each card and deck configuration.

Conclusion

It is crucial that a designer understands how he builds roles in his designs, and – almost as importantly – how he communicates them to players. While this discussion is just preliminary, the concept of “roles” as a design tool is extremely powerful and should not be ignored. One step further than that, game design as a communication situation is an even more powerful concept that provides us with a different perspective on travelled ground. What more new conceptual tools can we develop when we make use of this framework? Design as Communication can offer good insights on how we can impact the behavior, knowledge, attitude and emotions of players.

References

FRASCARA, J., 2007. Communication Design – Principles, Methods and Practice. Alworth Press.

ATKIN, A., 2006. Peirce`s Theory of Signs. Stanford Encyclopedia of Philosophy. Available from: http://plato.stanford.edu/entries/peirce-semiotics/ [Accessed 26 July 2011].

NORMAN, D., 1988. The Design of Everyday Things. Basic Books.

GIBSON, J., 1986. The Ecological Approach to Visual Perception. Psychology Press.

ZIMMERMAN, E. AND SALEN, K. 2008. Rules of Play: Game Design Fundamentals. The MIT Press.

KOSTER, R., 2005. A Theory of Fun for Game Design. Paraglyph Press.

ADAMS, D.,2002 . The Ultimate Hitchhiker's Guide to the Galaxy. Del Rey.

VENTURELLI, M., 2009. Space of Possibility and Pacing in Casual Game Design - A PopCap Case Study. Available here. [Accessed 8 August 2011].

While it is common to believe that playing a game “well” means excelling at the last step of decision-making, it is interpretation that provides the basis to work with. That is, a wrong interpretation of the game state and/or of the space of possibility may lead the entire process towards frustration and “bad decisions”. The first two steps, then, are more interesting for our discussion, as they form players’ mental model [Norman 1988] of the game. A mental model is a “thought reconstruction” of how something works in the real world, built by experience, training and instruction [Norman 1988] – a model built by semiosis.

It is pertinent, then, to discuss game state and space of possibility a little more in-depth. The game state is clearly and concisely explained in “Rules of Play”: