A Comprehensive Game Design Methodology
From First Ideas to Spectacular Pitches and Proposals

The content of this website is licensed under a Creative Commons CC BY-NC-SA 4.0 Attribution–NonCommercial– ShareAlike 4.0 International License. You can freely use and share everything non-commercially and under the same license as long as you attribute it to and link to:

J. Martin | |

However, you can also buy the Ludotronics PDF edition
for an unreasonably moderate price at DriveThruRPG.
Learn here about the five excellent reasons to do so!

Why DriveThruRPG? It’s the largest tabletop RPG download store and you’ll probably end up buying much more than just your copy of Ludotronics. Which would benefit all game designers!

Why not Amazon? Ludotronics isn’t well-suited for the Kindle format. And at €14.99, Amazon’s cut amounts to €9.75. Well, no.

More to read: My papers at Research Gate, my blogs at between drafts and just drafts.

Level One: Integral Perspectives I

Process Phase Level One


Now that you’ve worked hard to create and polish a viable game concept and trace out its principal premises as well as its marketability, you will proceed to sketch out everything that makes a game, from its motivators to its rewards, from its rules to its dramatic structure. It’s here where you will transform your original idea of a game into your vision of an interactive playing experience.

up | down

Beat 1. Experience

But wait—what does interactive playing experience mean in the first place? According to books, papers, interviews, and the internet in general, interactive playing experience means just about everything and just about nothing. Descriptions and definitions seem to be ranging from the theoretical to the practical and from the sweeping to the particular.

To make this term more useful and pave the way for meaningful, deliberate design decisions, let’s start with breaking down “experience” into the constituents that are applicable to our context. On one side, we have experience as expertise: developing skills, knowledge, understanding, and attitude toward mastery and competitive performance through motivational involvement. On the other, we have experience as awareness: perceiving ongoing events with and through emotional involvement.

As you can see, this presents a formidable obstacle to designing any kind of playing experience. We can’t design motivational and emotional involvement toward expertise and awareness directly. And even if we could, players should or indeed must be allowed to create their own experiences in games, which are not necessarily the experience the game designers had in mind.

Fig.4.1 Playing Experience (Definition)
Fig.4.1 Playing Experience (Definition)

Thus, playing experiences can only be designed indirectly, by designing game events that facilitate a certain range of possible experiences. From there, great design can make it happen that most playing experiences fall within the game’s intended range.

At its core, therefore, the playing experience is about player motivation (toward expertise) and player emotion (toward awareness). In other words:

  • Playing experience follows player motivation and player emotion.

Now we have to make this interactive, not to forget. Within the Ludotronics paradigm, player motivation and player emotion are methodologically tied to the four territories through the just-right amount of challenge (Interactivity), compelling aesthetics (Plurimediality), emotional appeal (Narrativity), and meaningful choices (Architectonics). Together, they comprise the interactive playing experience model of the Ludotronics paradigm.

Fig.4.2 Interactive Playing Experience Model
Fig.4.2 Interactive Playing Experience Model

Each territory, together with its motivational aspects, will have its own day in the sun during the Process phase. But motivation and emotion as such are the two integral perspectives that you, as a game designer, need to know inside out, so we will discuss them in great detail in the following two beats. Both come with a few tough bites of theory. But chewing your way through them and digesting them with determination will pay off in practice later.

up | down

Beat 2. Motivation

In the Procedure phase in Level Three: Tracing the Goal, we created two sets of motivational drivers for the desire-driven goal: simple, specific, and promising on one side, and excite, achieve, and become on the other. We can leave the first set behind now, as it is focused on attracting attention and marketing in general. It’s the second set that is deeply involved in creating motivation. It will follow us throughout the Process phase and eventually weld this first level to the final level, Level Six: Integral Perspectives II, through a wider context shared by motivation and rewards.

In this level, the second set’s three motivators constitute the three floors of what we will call the motivational building, with excite as its first floor, achieve as its second floor, and become as its third or top floor. Excite represents the first floor because it’s hard to become motivated if you’re not excited. On this floor resides the interactive playing experience model we built in the preceding beat. Achieve represents the second floor. This floor is where the player action is and where the rubber hits the road. It’s populated by a balanced cocktail of activities, the player activity setup, that we will introduce in a minute. Become, finally, represents the third floor. It holds all the excitement and all the achievements together as a highly desirable meta-motivation toward personal growth. On this floor lives the player motivation model, which we will construct later in this beat.

Fig.4.3 The Floors of the Motivational Building
Fig.4.3 The Floors of the Motivational Building

Let’s check all three floors of our building and say hello to the tenants.

On the first floor, as mentioned, reside the four elements of our interactive playing experience model: the just-right amount of challenge, compelling aesthetics, emotional appeal, and meaningful choices, each tied to one of the four territories.

The second floor of the motivational building is populated by a set of activities that we will call the player activity setup. This set is a cocktail of tasks that, if balanced in the right manner, makes these activities attractive, captivating, and worthwhile even though they are essentially work. For the ingredients of this cocktail, we will follow what we feel is the most convincing and comprehensive set of player activities brought forward so far, the seven types of work in games as established by Jane McGonigal in Reality Is Broken. These ingredients are high-stakes work, busywork, mental work, physical work, discovery work, team work, and creative work. We will meet these seven ingredients again in Level Two: Interactivity, where we will apply them to a game loop example, and in Level Six: Integral Perspectives II, where they will contribute to a game’s reward system.

The third or top floor needs a more thorough introduction and a few remarks on research.

There exist all kinds of motivational theories, relating to needs, expectancy, arousal, and so forth, some of them stronger on the empirical side than others, some better at making testable claims or accurate predictions than others, but on the whole they’re too broad for our purposes. In game-based learning research, two models are often used that are practical, applicable, and have a strong empirical heft. One of these is a learning motivation model, the other a work motivation model. Let’s have a look at both.

The motivational model for learning is based on a motivational theory that grew out of research on intrinsic and extrinsic motivation, mainly developed by Richard M. Ryan and Edward L. Deci, and applied to game design by Scott Rigby and Richard M. Ryan. (It’s called self-determination theory, but really, it’s a model.) Its three primary building blocks are autonomy, competence, and relatedness.

The other model, the motivational model for work, is also based on research on intrinsic and extrinsic motivation, put into shape and brought to public attention especially by Dan Ariely and Daniel H. Pink. In its most generalized form, it also boils down to three primary building blocks: autonomy, mastery, and purpose.

Autonomy is shared by both models. Mastery and competence as a pair are somewhat lopsided, and we will deal with that soon. Then there’s relatedness in one model and purpose in the other.

Relatedness means exactly what you think it means: the need to connect with others, to take care of others and be taken care of when things go wrong. Curiously, it’s absent from the work motivation model, which makes you wonder. Not only is relatedness fundamental to cooperation, collaboration, and cocreation, which in turn are fundamental to mastering any substantial challenge in our time. Relatedness also supports the willingness to take risks, entrepreneurial and otherwise, not least because relatedness is also vital in case you fail. While we shouldn’t put failure on too high a pedestal, it’s certainly an opportunity to learn from mistakes, to see what works and what doesn’t, and to push forward with better informed decisions. Which is equally true for leading a startup as it is for playing a game!

On the side of the motivational model for learning, purpose is missing. This has to do with a set of underlying—and indeed quite reasonable—assumptions that subordinate purpose to autonomy. In a nutshell: according to the self-determination model, intrinsic aspirations and purposes are experienced as meaningful and promote well-being because they are products of autonomous decisions (driven by curiosity, creativity, enjoyment, the search for knowledge, and similar). Now, while this is theoretically sound within its framework, some evidence suggests that the three motivational factors from self-determination theory might be incomplete in our particular context; see, for example, “Applying the Self Determination Theory of Motivation in Games Based Learning” by David Farrell and David Moffat.

Thus, the Ludotronics player motivation model will employ four primary building blocks instead of three, and each of them will come with two aspects that augment each other through essential characteristics.

The first building block consists of autonomy and agency. These are not the same—you can have one without the other. The player might have autonomy to make decisions, but to have agency, these decisions must also make a difference. Vice versa, the player can have agency and make a difference with their actions, but lack autonomy because they’re not calling the shots.

The second building block consists of mastery and, in lieu of competence, performance. Mastery covers aspirations to become the best at a task, or at least to become as good as one possibly can. Performance covers aspirations to perform better than others—in other words, to win. (We will look at mastery and performance goals in more detail in Level Six: Integral Perspectives II.)

The third building block consists of relatedness and community. Relatedness covers everything already mentioned: the need to connect with, and take care of, others; the willingness to take risks; and the opportunity to learn from mistakes. But relatedness does not necessarily include community. You can have great relationships with mutual support and everything without building or growing a community. Importantly, it’s not relationships but communities that develop distinct cultures. Not always in a good way, mind—gaming communities in particular are often riddled with problems like exclusion, discrimination, and toxic behavioral norms. All this is part of the package, though, and that’s what the term community brings to the table.

The fourth building block, finally, consists of purpose and goal. While more research needs to be done to test this hypothesis, observations indicate that the absence of purpose might be the factor that causes expectations to fail when self-determination theory is applied to game-based learning or game design in general (research pending). In contrast to purpose, the complementary term goal refers to clear-cut goals that can be set, followed, and achieved. This comprises not only the game-driven goal, but all three parts of the holistic goal including the design-driven goal and the desire-driven goal, and all kinds of self-set goals that players might create along the way.

To sum it up, the four building blocks of our player motivation model are autonomy/agency, mastery/performance, relatedness/community, and purpose/goal.

Fig.4.4 The Motivational Building
Fig.4.4 The Motivational Building

Now let’s proceed to how this translates to learning.

Research into game-based learning explores how and what we can learn by playing games. Games are exceptionally good at teaching anything and everything. So good, actually, that we can say that motivation to play equals motivation to learn. To play a game well, we have to overcome challenges, and to overcome challenges, we have to learn. All the time. That’s the nature of challenge. And what do you know—the widely familiar concept of flow, originally developed in the 1970s by the psychologist Csíkszentmihályi Mihály and subsequently expanded upon by himself and others, describes exactly that: the state of flow is the completely focused motivation to learn and perform through constant challenge.

The concept of flow, which we will examine more thoroughly in Level Two: Interactivity, is greatly preferable to the non-concept of “immersion,” a favorite staple in game design lore, together with its perfectly vacuous cousin “fun.” If you detach immersion from the concept of flow, you’ll be left with the conceptual equivalent of candy floss—fluffy and tasty and sticky as a concept but if you try to sink your teeth in it there’s essentially nothing there. You can be “immersed” in any media, from a blazingly high-tech virtual reality system at a cutting-edge research facility down to a cheap, worn-out paperback edition of a romance novel you read on your short bus ride to town. That’s not very helpful. So let’s stick with flow, which is much more sharply defined, and whose inner workings we will examine, as mentioned, in Level Two: Interactivity in greater detail. Flow is what holds our motivational building together: from the territory-based interactive playing experience model with its just-right amount of challenge, compelling aesthetics, emotional appeal, and meaningful choices to the player activity setup with its seven classes of work to the player motivation model with its four building blocks autonomy/agency, mastery/performance, relatedness/community, and purpose/goal.

Fig.4.5 Motivation to Play (Definition)
Fig.4.5 Motivation to Play (Definition)

up | down

Beat 3. Emotion

Now that we have the keys to our motivational building, let’s build a model for player emotion right in front of it that we will call the emotional landscape. As with motivation, the number of theories and models about emotions is gargantuan—how to categorize emotions, for example, or how they compare to each other, or how they relate to personality, temperament, and mood. Science offers us a good number of sharp knives to dissect emotions for evolutionary, neurological, sociological, or psychological research purposes, to name a few. But we will pick the comparatively crude box cutter knife of dramatic theory, notably Aristotle’s Poetics, to carve out a small number of equivalently crude categories. We will do that because it suffices for our expressed purpose of designing games. Yet, we need to remind ourselves from time to time that dramatic theory is only a modest chamber within the vast and splendid edifice of scientific research into emotions and emotional experience.

In our first operation, we make just one deep cut. On one side of the cut, we have all the emotions that can be triggered in the player by visual, auditory, and kinesthetic means directly, i.e., through images, sounds, melodies, and movement. Without any claim to completeness, these comprise:

  • Fear, joy, happiness, amusement, awe/wonder, sadness, sorrow, melancholy, nostalgia, anxiety, terror, dread, disgust, nausea, revulsion, and similar.

On the other side of our cut, we have all the emotions that cannot be triggered in a player by visual, auditory, or kinesthetic means alone. These comprise, again without any claim to completeness:

  • Pity, compassion, worry (for others), love, contempt, anger (not the kind directed at faulty controls), outrage, hate, admiration, envy, jealousy, resentment, grief, bitterness, and similar.

We can call the first group of emotions the “fear class” and the second group of emotions the “pity class.” And, mirabile dictu!, it’s these two classes that we find in the sixth chapter of Aristotle’s Poetics and his definition of Greek Tragedy! There, Aristotle describes Greek Tragedy as composed of certain elements and characteristics like imitation of action and artistic adornment, and he closes with the statement that Greek Tragedy, roughly, “effects through pity and fear the catharsis of these emotions.” Now, much has been written about the catharsis hypothesis, a concept that has so far resisted—or evaded—verification or falsification. But catharsis is not of immediate interest to us. Considerably less has been reflected upon the question whether this final part of Aristotle’s definition continues to describe what Greek Tragedy objectively is and does to earn its name—in other words, the effect it should have—or if Aristotle intended to state what Greek Tragedy is for—in other words, its purpose. A good argument can be made for the latter, but it would deviate too far from the matter of this treatise. What really interests us right now is Aristotle’s choice of pity and fear. Why these two? There’s a well-established answer to that, deduced from Aristotle’s definition of the tragic hero as a human being who must not be “superhuman” but have a flaw that makes the hero human and brings about the hero’s downfall. Within this context, the audience pities the hero and fears that they could face a similar fate themselves someday. This defines the difference between our two classes. Emotions from the fear class are directly experienced through identification with a situation we could see ourselves in while emotions from the pity class are indirectly experienced through identification with the character or characters by way of empathy mechanisms (possibly mirror neurons, but that’s still a matter of debate).

This has repercussions for the way we design emotions in games. We can mess up spectacularly in two ways when we fail to recognize that situation and character are different loci of identification with different types of emotion and different types of experience—a direct experience by seeing ourselves in the game world on the one hand, and an indirect experience by empathizing with the game world’s characters on the other.

It was dramatically and narratively highly developed interactive games, specifically, that messed this up and introduced a “locational” virus into these identification patterns that put everything in disarray. To isolate this virus, we need to undertake a detailed anamnesis. Based on the definitions above, what’s been going on in non-interactive media for about three thousand years is this:

  1. The dramatic situation the audience can see themselves in (fear class).
  2. The characters the audience “identifies” with (pity class).

Now, in interactive games, our most popular strategy to create and convey emotions from the fear class consists of operating directly on the player with a sophisticated visual, auditory, and motion palette and also, particularly in console games, kinesthetic effects like vibrations and pulsations. This is particularly true for horror games, among them Silent Hill, The Suffering, or Dead Space. There, players experience these emotions directly, not indirectly through empathy with the player character or any other game world character, let alone identification.

So far, this is not too different from Greek Tragedy; especially Aeschylus could frighten the audience out of their wits through well-placed effects. Now, if this fear class were all that’s happening in interactive games, then there’d be no problem. But as soon as we create a dramatically complete game that also conveys emotions from the pity class, we have created a problem factory.

Our premise is that the player identifies with their player character. Therefore, in the tradition of Greek Tragedy and because it’s just too obvious, we are tempted to convey the emotions from the pity class by letting things happen to the player character. But is our premise that the player identifies with the player character really true? And if it were true, would that help?

Probably not. To start with, the term player character itself already suggests a measure of identification that is misleading because it obscures the fact that we merely deal with a player-controlled character. There’s a difference between controlling a ball or being a ball, and that’s also true for everything from cars to cybernetic super soldiers.

But there’s more. What the player identifies with, first and foremost, is the player-avatar as a representation of the player’s actions. This is not the same as identifying with a player-controlled character as a character. These are completely different mechanisms, which is and isn’t obvious at the same time.

To boot, a major source of confusion is our indiscriminate use of the term “identify with.” What we mean by that when talking about characters in non-interactive media, like novels or movies, is “empathize with,” and that’s what everybody understands, at least implicitly. But when it comes to player avatars in interactive games, we tend to forget that. Here, we often use “identify with” as if it were indeed about identification, not about empathy. Which, for game design purposes, is the opposite of helpful.

Thus, for a better understanding, we need to ditch the “identify/identification” terminology for good. It forces us to see more clearly that there are not two, but three loci of emotional involvement in interactive games:

  1. The dramatic situation the audience can see themselves in (fear class).
  2. The characters the players can empathize with (pity class).
  3. The player avatar that represents the players’ actions.

The third locus—with a class of emotions of its own that we will examine below—is exclusive to video games and comparable interactive media where it generates a whole spectrum of headaches. What’s generally taken for granted is that (2), the character the player empathizes with, and (3), the player avatar that represents the player’s actions, are more or less in the same ballpark, so that establishing one automatically establishes the other. Alas, this is entirely not the case. Not only are these classes conceptually distinct, they also trigger different emotions. For the third class, again without any claim to completeness, these comprise:

  • Triumph, relief, jubilation, satisfaction, pride, gratitude, confidence, hope, guilt, remorse, shame, embarrassment, disgrace, regret, humility, disappointment, frustration, despair, and similar.

In non-interactive media, these emotions can only be experienced as part of the pity class through empathizing with the characters. But in video games and comparable interactive media, where players have agency and can make decisions that have consequences, it is possible to trigger these emotions in a way that would be impossible to trigger with novels or movies: the player can experience guilt, remorse, shame, regret, pride, triumph, relief, etc. directly as a consequence of their own actions and decisions. To stress it again, this is unique to media that allow for agency, with video games as the most prominent example.

If Aristotle lived today and looked at video games as morosely as Classical Greeks looked at pretty much everything, he’d probably extend his original definition to fear, pity, and guilt. It’s tempting to do just that. But to give this class of emotions a more positive spin, we will instead call it the fiero class:

  1. The dramatic situation the audience can see themselves in (fear class).
  2. The characters the players can empathize with (pity class).
  3. The player avatar that represents the players’ actions (fiero class).

Fiero is a concept that was introduced to the game design community by Nicole Lazzaro through her GDC and AIGAdesign presentations. It was quickly spread by others, deservedly, and became well-established as a term for the overwhelming feeling of accomplishment that includes triumph, pride, relief, and such, when players have overcome a major obstacle, triumphed in the face of adversity, won a huge victory, and similar. Most of these emotions are acknowledged and reflected within game worlds by design through rewards, acknowledgment by NPCs, and many other things. But not all of them, or always. For example, as David Perry and Rusel de Maria observe in David Perry on Game Design, the player emotion of pride often fails to be reflected within game worlds.

But this coin too has two sides. In the same way that our fear and pity classes also comprise positive emotions, the fiero class comprises negative emotions. Among them are shame, guilt, regret, or disgrace that occur when a player fails to overcome an obstacle; suffers defeat or takes major losses; or makes life worse for their player character, their non-player companion characters, or other players’ characters in coop or multiplayer games or MMOs.

Based on our understanding of these three loci of emotional engagement, we can now tackle the interference problems that occur between the pity class and the fiero class when they are confused with each other in ways indicated above. Depending on mechanical aspects, some dramatically complete game types are traditionally more vulnerable to this confusion than others. Let’s have a look at three major types.

Dramatically complete games with “detached” player characters, as we will call it, are usually unproblematic in this regard. Consider adventure games with protagonists that represent the player’s actions as player avatars without being supposed to be identified with, like Syberia’s Kate or Grim Fandango’s Manny, or, to a certain extent, Tomb Raider’s Lara Croft. Detached player characters have their own agendas, they often decide for themselves, and they have their own life. Kate or Manny or Lara represent the player’s actions as player avatars, they don’t represent the player as player. This leaves the emotional involvement structure largely intact. Players can see themselves in the dramatic situation (fear class), and they can empathize with the player character (pity class) who also represents the players’ actions (fiero class).

Then there’s the type of dramatically complete games where the player is indeed supposed to be the player character in a way defined by John Romero. According to the latter’s famous forum post, there never was a name for the Doom marine “because it’s supposed to be YOU.” Such games do offer stories, even if rudimentary. But they’re not completely dramatically complete because they don’t offer all three emotional classes: rarely is the player supposed to feel pity, hate, love, or grief. The pity class remains empty, or undefined, and that’s why the structure of emotional involvement remains largely intact again. Players can see themselves in the dramatic situation (fear class), and the player character represents the players’ actions (fiero class). That’s it. Which is similar to, but not identical with, games that are not dramatically complete, from Pong to Forza Motorsport to Unreal Tournament. Not only are the players’ actions represented by the player avatar for the fiero class’s range of emotions in these games, they also offer a residual situational in place–feeling at the point of one’s paddle, wheel, or hand-&-gun that is able to evoke a rich subset from the fear class’s range of emotions. But, like their not-quite dramatically complete Doom type cousins, they do not offer emotions from the pity class.

It’s the third major type of dramatically complete games that gives us trouble. These are games where the pity class is intact and the player character is supposed to represent the player, not just the player’s actions. With this combination, two incompatible identification routes collide: pity (identification via empathy) and agency (identification via representation).

Let’s illustrate this dilemma with a stock example. In a historical game, the player character must choose between her fiancé, whom she loves, and her dream career. When this character chooses one of these options herself, the player can empathize and pity her for sacrificing either her love or her career. But that no longer works when it’s the player who makes this decision. You can’t possibly empathize with yourself, because that makes no sense whatsoever. And you can’t feel pity for yourself, only self-pity, which is a completely different thing. In other words: with such a player character, you can’t have empathy and agency at the same time!

Yet, many games want to have their cake and eat it too by endowing players with agency and with emotions from the pity class simultaneously. It’s possible, to a certain degree, but only through an approach that contributes to video game narratives’ bad rap. It’s an approach that employs clichés as lowest common denominators, with easily recognizable emotions triggered by easily recognizable situations which are so familiar to us that players can “feel” these emotions without really feeling them, without having to step in anyone’s shoes, without having to empathize with anyone in any way. This is what makes dramatically complete games often so poor. It’s not simply “poor writing,” no. That could be remedied. It’s a dilemma, and it is structurally related to the time-honored dilemma between dramatic structure and player agency, to be discussed in Level Five: Architectonics (which postulates that you can’t have a dramatic roller coaster ride experience, similar to books or movies, and true player agency at the same time).

Does this mean that we cannot possibly accomplish a dramatically complete game that is emotionally meaningful and not poor? Fortunately, this isn’t the case. Two approaches exist, at least. One is clever tinkering in the agency department, the other a well-established approach in the empathy department that focuses on the use of NPCs. To see what great design can accomplish by combining these two approaches, let’s have a brief look at the games from the Mass Effect trilogy (minus the ending).

Clever tinkering in Mass Effect begins with almost unlimited player character customizability. On top of appearance and a wide range of character classes, abilities, and even personal backgrounds to choose from, which all affect the game in one way or another, players can indeed personalize their player character’s personality. The player can “own” the player character and, to a degree, project their own values, traits, and emotional make-up by choosing the player character’s moral values through dialogue options and decisions. This gives the player a substantial amount of autonomy. Different things happen after different decisions, which are frequently less about what to do, but how to do something, which gives the player also a substantial amount of agency. (The difference between what and how as a basic task design principle will occupy us later in Level Five: Architectonics.)

Certainly, it’s comparatively crude, and the choices are limited. But it’s an ingenious way of providing players with the option to project at least some of their preferred values and characteristics onto the player character’s behavior, and have the game respond accordingly. It’s still a simulation of agency, not true agency. But we don’t yet have the technology to parse, interpret, and respond to the player’s emotional and moral behavior consistently in real time. Right now, it’s as close as we can get.

But, as clever as this approach is, outside the agency department it has its limitations. In the empathy department, it still isn’t enough to carry the player sufficiently deep into the rich palette of emotions that make up the pity class. That’s where the second approach kicks in. There’s an old quip where a seasoned, experienced character boasts they’ve “forgotten more things than most people ever learned,” and this second approach fits that quip as it is one of the oldest tricks in the book. It’s about letting dramatic turns or events with high emotional impact not happen to the player character but to hero NPCs, major non-player characters, to which the player character has become sufficiently attached. (Hero NPC is a term adapted from Chris L’Etoile, to whom we will get back in Level Five: Architectonics.) Originally, this served as a safety feature to protect players from emotional impacts that would yank them out of their comfort zone. That’s still important, but we can also use it to serve our purposes here. Remember, you can’t empathize with yourself. You can only empathize with others, and that works better and more reliably when you empathize with someone who really means something to you. The games from the Mass Effect trilogy are chock-full of hero NPCs one can care about deeply, providing the player with a rich subset of emotions from the pity class.

Combining both approaches, the empathy/agency dilemma almost disappears. The player can enjoy both at the same time: the emotions from the pity class by empathizing with the fates of hero NPCs and player agency enriched by a limited amount of personalized emotions.

Fig.4.6 The Emotional Landscape
Fig.4.6 The Emotional Landscape

There’s one aspect we haven’t talked about yet but should, particularly in the context of our Mass Effect example: romance options and love as an emotion from the pity class.

Do players really fall in love with Garrus, Samantha, Liara, Steve, or Tali, or characters from any other video game in general? Well, there are fringe cases like “Japanese man falls in love with/marries video game character.” But whatever the future has in store for us in terms of pygmalionesque AI, at this point in time the general answer would be “no.” What players feel toward video game characters is not love, but strong emotional attachment. This isn’t weird. Video game characters aren’t special in that regard. Humans can, and do, develop deep emotional attachments to everything from dolls and teddy bears and plush bunnies to guns and hats and automobiles. In video games, it’s this capacity that makes it possible to project the emotional potential of an in-game event, including romantic love between the player character and non-player characters, into the mind of the player without having to make the player “truly” fall in love with an organized collection of pixels.

For games that aim at being dramatically complete, that’s the benchmark—at least for the moment. New technologies and applications like machine learning will certainly give us more options in the long run. But right now it’s a good idea to learn from approaches that already exist, and develop these approaches further through improving and refining them.

But let’s not draw the wrong conclusions. Our Mass Effect example notwithstanding, it’s not about complex technologies or staggering budgets. It’s about designing and controlling all three loci of identification at all times throughout your game: the dramatic situation players can see themselves in; the characters players can empathize with; and the player avatar that represents the players’ actions. These three loci define the player’s emotional landscape. If you want to design a dramatically complete game, controlling this landscape is as indispensable for your first indie title as it is for your third AAA release.

up | down


Together, the motivational building and the emotional landscape define what your player will experience and how they will experience it. These models, or frameworks, enable you to make design decisions for numerous aspects across the upcoming levels. As a game designer, you really need to know a lot about motivation in general, and motivation with regard to your target audience in particular. If you want to design a dramatically complete game, you also need to know a lot about emotions in general, and emotions with regard to your target audience in particular. What motivates a player for a certain course of action and what doesn’t? What elicits a specific emotion and no other? These are questions that you’ll have to answer all the time.

One last point. Design decisions for each class from the emotional landscape are loosely attached to design decisions in one of the four territories, as you might have guessed already from the landscape’s color scheme. Fiero interconnects strongly with the Interactivity territory, fear with the Plurimediality territory, and pity with the Narrativity territory. And in the fourth territory, Architectonics, you will pull the dramatic strings that hold all three together.