MDA Framework: An S-Tier Behavioral Designer’s Guide

Ask ten game designers what makes a game good and you will get ten different answers, most of them circling the same useless word: fun. The problem is that “fun” tells you nothing you can build. You cannot open a code editor and type fun. You cannot put fun on a spec sheet or assign it to a sprint. For decades the industry shipped games the way short-order cooks shipped breakfast, by feel and by reputation, with no shared language to explain why one combination of rules produced delight and a nearly identical one produced boredom.

In 2004, three designers put a stop to that. Robin Hunicke, Marc LeBlanc, and Robert Zubek published a short paper that gave the field its first clean vocabulary for taking a game apart. They called it MDA, for Mechanics, Dynamics, and Aesthetics. The idea was almost embarrassingly simple: every game can be read on three layers, from the rules the designer writes, to the behavior those rules produce in play, to the feelings that behavior creates in the person holding the controller. Name the layers, and suddenly you can argue about a design without waving your hands.

Here is the part that still matters twenty years later, and the part most summaries miss. MDA’s deepest insight is not the three letters. It is the direction. The designer builds a game from the bottom up, mechanics first, hoping for the right feeling at the top. The player meets the game from the top down, feeling first, only later noticing the rules underneath. The two of you are reading the same object from opposite ends, and the entire craft of design lives in that gap. This guide breaks down what Hunicke, LeBlanc, and Zubek actually proposed, the eight specific kinds of “fun” they named, where the framework quietly falls apart, and how it connects to the motivational engine it stopped just short of describing.

Speed Run Notes

• MDA splits any game into three causally linked layers: Mechanics (the coded rules), Dynamics (the behavior that emerges in play), and Aesthetics (the emotional response in the player).
• The core insight is directional: designers build Mechanics → Dynamics → Aesthetics, but players experience Aesthetics → Dynamics → Mechanics. You read the same game from opposite ends.
• “Fun” is not one thing. LeBlanc named eight: Sensation, Fantasy, Narrative, Challenge, Fellowship, Discovery, Expression, and Submission. Naming which one you are building is half the design job.
• You only directly control Mechanics. Dynamics emerge, and Aesthetics can only be coaxed. Most design failures are someone trying to command a layer they can only influence.
• MDA’s limit is that it names the target emotion but not the motivational cause. It tells you to aim for Challenge; it never explains why a human wants challenge.
• That gap is exactly where Octalysis picks up: MDA’s aesthetics describe the felt experience, the 8 Core Drives describe the engine producing it. The two frameworks stack rather than compete.

Table of Contents

Author Credibility: Yu-kai Chou

Yu-kai Chou created the Octalysis Framework after studying gamification since 2003 — years before the term entered mainstream vocabulary. As a Human-Systems Architect & Behavioral Designer, his framework has been applied by LEGO, Microsoft, Porsche, Coca-Cola, Salesforce, and MrBeast, impacting over 1.5 Billion Users.

Chou has taught the Octalysis methodology at Harvard, Stanford, Yale, Tesla, Google, BCG, and IDEO.

His work has been cited by Harvard, Stanford, MIT, Forbes, Wall Street Journal, Wired, US Department of Energy, NIST, NSF, NCBI, US Department of Education, ClinicalTrials.gov, and Google Scholar — with 3,700+ more academic publications. Explore his books here.

What Is the MDA Framework?

The MDA framework is a way of reading any game as three connected layers. Mechanics are the rules and components a designer writes: the cards, the dice, the hit points, the cooldown timers, the exact code that decides what happens when a button is pressed. Dynamics are what those rules actually do once a real person starts playing: the bluffing that emerges from betting rules, the camping that emerges from spawn points, the rich-get-richer spiral that emerges from Monopoly’s property system. Aesthetics are the emotional responses the dynamics produce: tension, triumph, belonging, surprise, the particular flavor of enjoyment a player walks away with.

The three are causally linked, like a chain. Mechanics give rise to dynamics. Dynamics produce aesthetics. Change a rule at the bottom and the effect ripples all the way up to how the game feels, often in ways you did not predict. That last clause is the whole reason the framework exists. In most products you build, you can roughly see the line from feature to outcome. In games, the rules collide with each other and with an unpredictable human, and the result is genuinely hard to foresee. MDA gives you a map so you can at least reason about where to push.

There is one more idea baked into the original paper that sounds abstract but changes how you think. Hunicke, LeBlanc, and Zubek argued that games are more like artifacts than media. A movie streams its content at you in a fixed order. A game’s content is its behavior, the space of things that can happen when you interact with it. The pixels and music are surface. The real substance is the system underneath, the set of possibilities the rules open up. Once you accept that, you stop designing scenes and start designing systems, which is exactly the shift MDA was built to support.

The Paper That Named It

MDA did not arrive out of nowhere. It grew out of the Game Design and Tuning Workshop that Marc LeBlanc ran at the Game Developers Conference in San Jose every year from 2001 to 2004. LeBlanc had been teaching working designers how to take games apart and rebuild them, and the three-layer model was the spine of that teaching. In 2004, Hunicke, LeBlanc, and Zubek wrote it up as “MDA: A Formal Approach to Game Design and Game Research” and presented it at the AAAI Workshop on Challenges in Game AI. A short conference paper, written for an artificial-intelligence audience, became one of the most cited documents in all of game design.

The authors were not academics theorizing from the outside. They were practitioners. Robin Hunicke went on to produce Journey at thatgamecompany, one of the most emotionally resonant games ever shipped, and later co-founded the studio Funomena before becoming a professor at UC Santa Cruz. Marc LeBlanc had already worked on System Shock and Thief: The Dark Project, two games whose systemic depth is still studied, and he has spent the last decade as a principal engineer at Riot Games. Robert Zubek built large-scale social games at Zynga, co-founded the studio SomaSim, and later wrote Elements of Game Design for MIT Press. MDA was distilled from people who had to make the rent by making games people actually enjoyed.

That origin matters because it explains the framework’s character. It is not a grand unified theory and never claimed to be. The paper says plainly that there is no formula that tells you the exact mix of ingredients that produces fun. MDA is a working vocabulary, a set of handles a team can grab so they can argue productively about a design instead of talking past each other. It was meant to be used in a room, on a deadline, by people trying to fix a game that was not working yet.

The Three Lenses: Mechanics, Dynamics, Aesthetics

The authors described each layer as a lens, a separate view of the same object. Looking through one lens does not move the others, but the lenses are causally connected, so what you see through one tells you something about the rest. Here is each one in the detail a working designer needs.

Mechanics: the rules you actually write

Mechanics are the most concrete layer, the only one you touch directly. They are the actions, behaviors, and control mechanisms available to the player, plus the content the game is made of: the levels, the assets, the numbers. The mechanics of a card game are shuffling, trick-taking, and betting. The mechanics of a shooter are weapons, ammunition, and spawn points. The mechanics of golf are balls, clubs, sand traps, and water hazards. When you adjust a mechanic, you are turning a knob whose effects you can specify exactly. The cooldown is now four seconds instead of six. The deck has two more aces. This precision is why mechanics are where tuning happens.

Dynamics: the behavior that emerges in play

Dynamics are what the mechanics do once they meet a player and each other. Nobody designed bluffing into poker as a discrete rule. Bluffing emerges from the combination of hidden cards and betting. Nobody wrote a “camping” rule into a shooter. Camping emerges from the interaction of spawn points, sightlines, and weapon ranges. Dynamics are the run-time behavior of the system, and they are where the surprises live. This is the layer you cannot edit directly. You can only change the mechanics underneath and watch what new behavior appears, which is exactly why playtesting is non-negotiable. You are not checking whether your rules work. You are discovering what your rules actually do.

Aesthetics: the feeling the player walks away with

Aesthetics are the emotional responses the game evokes. The original paper is careful here, because “fun” is the trap. The whole point of the aesthetics layer is to retire that one mushy word and replace it with a more precise vocabulary: not “this game is fun” but “this game delivers tension, fellowship, and discovery.” Aesthetics are the felt experience, the reason a player keeps coming back or quietly uninstalls. And crucially, the aesthetics are the goal. Mechanics and dynamics are means. The feeling is the end. A designer who forgets that ends up polishing systems that work perfectly and move no one.

A worked example: fixing Monopoly

The original paper teaches the three layers through Monopoly, and the example is worth walking because it shows the chain in motion. Start at the aesthetics. The feeling Monopoly wants is tension, the drama of a close contest where any player might still win. Now look at the dynamics that actually emerge from the rules. Once one player owns a few monopolies, they collect rent faster, which lets them buy more property, which lets them collect even more rent. The rich get richer and the poor get poorer in a widening spiral. Long before the official end, the outcome is decided, and the other players are just rolling dice on a result they cannot change. The tension the game wanted has quietly drained out of the room.

Where did it go wrong? Not at the aesthetics, which were the right target, and not really at the dynamics, which are just doing what the rules tell them. The leverage point is the mechanics, the only layer you can edit. The paper offers two families of fixes. One is to flatten the runaway feedback loop directly, with subsidies that help lagging players or taxes that slow rich ones, calculated when passing Go or when a player’s holdings cross a threshold. The other is to shorten the game so the spiral never has time to fully run, by depleting cash over time so people spend quickly, or doubling monopoly payouts so the field separates fast and a winner emerges before everyone gets bored. Either way, you change a number at the bottom, then play again and watch whether the tension at the top returns. That loop is MDA in a single sitting.

Notice what the framework gave you. It did not magically produce the fix. It told you which layer to look at, in what order, so you stopped arguing about whether Monopoly is “boring” in the abstract and started adjusting the specific mechanics that drain its specific target emotion. That is the everyday value of the three lenses.

The Eight Kinds of Fun

If “fun” is too vague to build with, what do you replace it with? LeBlanc’s answer was a taxonomy of eight distinct aesthetics, often called the eight kinds of fun. It is not meant to be the final word, and the paper says so, but it gives you eight specific targets instead of one fuzzy one. Here is each, with the one-line definition from the original work.

1. Sensation is game as sense-pleasure: the raw delight of sights, sounds, and feel. The crunch of a good sound effect, the bloom of a particle explosion, the tactile snap of a well-made board game piece. This is the aesthetic you feel in the first three seconds, before you understand a single rule, which is why it does so much work in trailers and store pages.

2. Fantasy is game as make-believe: stepping into an imagined world and an imagined role. Being a wizard, a city planner, a getaway driver, someone you are not. Fantasy is what lets a spreadsheet of resource numbers feel like running a kingdom; the fiction reframes the math as meaning.

3. Narrative is game as drama: a story that pulls you forward, the need to know what happens next. The key move in MDA is that narrative is not only scripted cutscenes. The most gripping stories in games are often emergent, the come-from-behind sports match or the heist that nearly went wrong, generated live by the dynamics.

4. Challenge is game as obstacle course: the urge to master something hard. This is the aesthetic most casual observers mistake for the whole of games, and it is only one of eight. Tuned well it produces flow; tuned badly it produces either boredom or rage, which is why this one aesthetic has an entire theory of its own.

5. Fellowship is game as social framework: belonging, cooperation, the pleasure of being part of a group with shared stakes. Guilds, teams, the kitchen-table energy of a board-game night. Fellowship is the stickiest of the eight, because people will tolerate a mediocre game to stay with the people in it.

6. Discovery is game as uncharted territory: the pull to explore, to find what is over the next hill or behind the locked door. Discovery rewards a designer who hides things on purpose and trusts players to go looking, rather than signposting every path.

7. Expression is game as self-discovery: leaving your mark. Building a base, designing a character, decorating a home, making something that is recognizably yours. Expression is why user-generated content outlives the campaign that shipped with the game; players keep returning to a canvas long after they have finished a story.

8. Submission is game as pastime: the comfortable surrender to a game as a routine, the cozy loop you return to without thinking. The daily crossword, the idle tap, the wind-down session. Submission is quietly the most commercially powerful aesthetic of the eight, because habit, not excitement, is what drives long-term retention.

The power of the list is comparative. The authors showed it by mapping four very different games. Charades runs on Fellowship, Expression, and Challenge. Quake runs on Challenge, Sensation, and Fantasy, with a heavy dose of competition. The Sims runs on Discovery, Fantasy, Expression, and Narrative. Final Fantasy stacks Fantasy, Narrative, Expression, Discovery, Challenge, and Submission all at once. Each game pursues several aesthetics in different proportions, and that blend is its fingerprint. The list does not tell you how to make a game fun. It tells you which fun you are aiming for, which turns out to be the harder and more useful question.

One honest footnote, because precision is the point. In the original examples the authors list “competition” as part of Quake’s appeal, even though competition is not one of the eight named aesthetics. They treated it as a higher-level pattern built out of dynamics rather than a base flavor. The eight-item list was always meant as a starting vocabulary, not a sealed periodic table, and that openness is both its strength and, as we will see, one of its weaknesses.

What Hunicke, LeBlanc & Zubek Got Right

Three things in MDA were genuinely new in 2004, and all three have held up.

The first is the separation of layers. Before MDA, design critique was a swamp of mixed levels, where someone would complain that a game “felt grindy” and someone else would answer by changing a damage number, with no shared way to tell whether they were even discussing the same problem. MDA gave teams a way to locate a problem. Is this a mechanics issue, a dynamics issue, or an aesthetics issue? That single question saves more design meetings than any other tool I know.

The second is experience-first thinking. By insisting that aesthetics are the goal and mechanics merely the means, MDA flipped the default workflow. Instead of starting from a cool feature and hoping a good feeling falls out, you start from the feeling you want and reason backward to the rules that might produce it. The paper calls this experience-driven rather than feature-driven design, and it is the same instinct behind every strong product, game or not. Decide what the human should feel, then build toward it.

The paper makes this vivid with a single concept stretched across three audiences. Imagine a game about a babysitter chasing a baby. Build it for toddlers and the target aesthetic is gentle Discovery, so the mechanics stay soft: peekaboo, simple chasing, surprised baby reactions, no real way to lose. Now aim the same premise at older kids and the target shifts toward Challenge and Narrative, so the mechanics grow teeth: time pressure to get everyone to bed by nine, a mess factor, multiple children with their own behaviors, levels that tell a story. Push it all the way to adults and reframe it as a stealth military simulation in the vein of Thief, where the aesthetic is Fantasy bordering on Submission. Now the mechanics explode into skill trees, enemy unit types, line-of-sight systems, and sophisticated opponent AI. Same core idea, three target feelings, three radically different rule sets. That is experience-first design in one example: the aesthetic you choose reaches all the way down and rewrites the mechanics.

The third is the honesty about emergence. MDA never pretends dynamics are designable. It tells you directly that you build mechanics, then dynamics happen, and your job is to predict and tune rather than command. The Monopoly example in the paper is the perfect teaching case: the rules create a runaway leader, dramatic tension collapses, and the fix is not to lecture players but to adjust the mechanics, with subsidies for the poor or taxes on the rich, then test again. That loop, build then observe then tune, is the actual rhythm of making things that involve real people. MDA was one of the first frameworks to put it on paper as a discipline rather than folklore.

Where MDA Falls Apart

MDA is a great map. Like every map, it leaves things out, and a few of the gaps matter enough that an entire cottage industry of successor frameworks grew up to patch them.

The aesthetics list is incomplete and a little arbitrary

Eight kinds of fun is a useful start, but it is not exhaustive and the authors knew it. Competition slips in through the side door in their own examples. Where does humor live? Schadenfreude? The specific satisfaction of optimization that drives spreadsheet-heavy strategy games? Designers have proposed longer lists ever since, which is a polite way of saying the original taxonomy was a sketch. If your design depends on an emotion the list does not name, the framework gives you no handle for it.

You cannot actually control the layer that matters most

The honesty about emergence is a strength, but it is also the framework’s central frustration. Dynamics are where the magic happens, and dynamics are precisely what you cannot author. You write mechanics and pray. MDA names this problem clearly but offers no method for crossing the gap, no reliable technique for getting from “here is the feeling I want” to “here are the exact rules that will produce it.” It hands you the right question and then steps back. That is not a flaw so much as an unfinished sentence, and it is the reason later frameworks tried to fill in the missing causal machinery.

It blurs the gap between intended and actual experience

MDA talks about the aesthetics the designer is aiming for as if hitting them were a matter of competence. But every player arrives with a different history, mood, and skill level, so the same dynamics land as triumph for one person and frustration for another. The framework has no clean way to distinguish the experience you intended from the experience the player actually had. This is the specific gap the DDE framework, Design, Dynamics, Experience, was built in 2017 to address, by separating intended design from lived experience and adding the transfer of feeling between the game world and the real world. The critique is real, and MDA’s own authors would likely agree the model was always a first draft.

What’s Really Happening: Emergence and the Two Directions

Strip MDA down to its load-bearing idea and you find two things, and the second is the one almost everyone skips.

The first is emergence. Mechanics are simple and local. Dynamics are complex and global. The gap between them is the same gap you see in a flock of birds, where a few dumb rules followed by each bird produce a swirling pattern no single bird designed. A game designer writes the per-bird rules and gets the flock as output. This is why games are unpredictable in a way movies are not, and why the only honest way to know what your rules do is to let real people loose on them and watch. You are not authoring outcomes. You are authoring the conditions from which outcomes emerge.

The second idea is the one that earns MDA its place in this library: the two directions. The designer builds from the bottom up. She starts with mechanics, watches the dynamics they produce, and hopes the aesthetics land. The player meets the game from the top down. He feels an aesthetic first, the tone and the mood, then starts noticing the dynamics that create it, and only the most curious players ever dig down to the mechanics underneath. Designer and player traverse the exact same three layers in opposite order.

That asymmetry is the whole craft. The designer’s deliverable lives at the bottom of the stack, and the thing she actually cares about lives at the top, and she can never touch the top directly. She can only reach it through the player’s own upward journey from feeling to behavior to rule. Every design mistake you can name is some version of forgetting this: tuning a number you love while the player feels nothing, or chasing a feeling with no mechanical path to produce it. Good design is the discipline of holding both ends of the chain in your head at once, building at the bottom while you imagine the experience at the top.

MDA vs Other Frameworks

MDA was first, and almost every game-design framework since defines itself partly in relation to it. Four comparisons clarify what MDA is and is not.

MDA vs the Octalysis Framework

This is the most important comparison, because the two frameworks are not rivals. They operate on different layers. MDA’s aesthetics tell you what emotional experience you are targeting. The Octalysis Framework tells you why a human would be motivated to pursue it. MDA says “aim for Challenge.” Octalysis says “Challenge satisfies Core Drive 2 (Development & Accomplishment), and here is how to build that drive into your system.” Put simply, MDA names the felt experience; Octalysis names the motivational engine producing it. We will stack them deliberately in the Octalysis section below.

MDA vs Bartle’s Player Types

Bartle’s Player Types sort players into Achievers, Explorers, Socializers, and Killers. MDA’s aesthetics describe experiences a game can offer. The connection is direct: an Explorer is chasing the Discovery aesthetic, a Socializer is chasing Fellowship, an Achiever is chasing Challenge. Bartle starts from the person; MDA starts from the experience. Read together, they let you ask which aesthetics your audience actually wants, then design the dynamics to deliver them.

MDA vs Flow Theory

Flow Theory describes the state of total absorption that happens when challenge and skill stay balanced. In MDA terms, Flow is a deep dive into a single aesthetic, Challenge, explaining the precise mechanical conditions that make it work. MDA gives you the wide vocabulary of eight aesthetics; Flow gives you the engineering manual for one of them. They fit together like a map and a zoomed-in street view.

MDA vs DDE and the successors

The Design, Dynamics, Experience framework, published by Walk, Görlich, and Barrett in 2017, is the most direct attempt to upgrade MDA. It keeps the three-layer spirit but splits the designer’s blueprint from the player’s lived experience and folds in the transfer between game and real life. Other academic efforts, including a 2021 push to turn MDA into a formal game-design ontology, try to make the vocabulary more rigorous. None has displaced MDA, which tells you something: a framework simple enough to teach in an afternoon often beats a more complete one nobody remembers under deadline.

It is worth naming the framework’s own lineage too, because MDA did not invent system-thinking about games on its own. It built on Doug Church’s late-1990s call for “formal abstract design tools,” a shared language for design, and it sits alongside Jesse Schell’s later idea of design “lenses,” where each lens is a question that reveals one facet of a game. MDA’s contribution inside that conversation was its insistence on the causal chain and the two directions: not just a list of questions to ask, but a claim about how rules, behavior, and feeling connect, and about how that connection looks different from the maker’s chair than from the player’s. That is the piece that made it sticky, and the piece every successor has had to reckon with.

MDA in the Real World

MDA was written for game studios, but the three-layer reading travels anywhere people build interactive systems.

Game and product design

The original use case is still the strongest. When a game feels off, MDA gives the team a diagnostic order: name the aesthetic you wanted, check whether the dynamics in playtesting are producing it, then trace the dynamics back to the mechanics you can actually change. It stops the meeting where everyone proposes random number tweaks and starts the one where you fix the right layer.

It also reframes the most useful question in a playtest. Inexperienced teams watch a tester and ask “did they like it?” That answer is an aesthetic, and it gives you nothing to act on. An MDA-literate team watches for dynamics instead: what did the player actually do, what behavior did the rules produce, where did they hesitate, optimize, exploit, or give up? Those observed behaviors are the bridge between the feeling you got and the rules you can change. The feeling tells you whether there is a problem; the dynamics tell you where it is; the mechanics are where you fix it. Run enough loops that way and design stops feeling like guessing and starts feeling like steering.

Gamified apps and software

A fitness app’s mechanics are streaks, badges, and point totals. Its dynamics are the behaviors those rules produce: the 11 p.m. workout to save a streak, the gaming of step counts, the quiet abandonment after a missed day. Its aesthetics are what the user feels: pride, pressure, guilt. Most gamification fails because teams ship mechanics, points and badges, while assuming the aesthetics will follow. MDA forces the harder question first: what should this feel like, and do our dynamics actually produce that?

Language apps make the trap concrete. A learning app’s mechanics are lessons, experience points, streaks, and leaderboards. Run them through MDA and you can predict the dynamics before you ship: some learners chase the streak so hard they tap through the easiest lesson each night just to keep the counter alive, which produces the aesthetic of anxious obligation rather than the Discovery and Challenge a learning product actually wants. The mechanic worked exactly as written. The dynamic it produced, streak-protection busywork, undercut the target feeling. A team reading the product through MDA catches that mismatch in playtesting and tunes the mechanic, perhaps by rewarding genuine difficulty instead of mere daily presence, rather than shipping a feature that technically functions and quietly trains the wrong behavior.

Education and training

An instructor designs mechanics, the assignments, grades, and deadlines. The dynamics are how students actually behave under those rules: cramming, gaming the rubric, helping or competing with each other. The aesthetics are whether learning feels like Discovery and Challenge or like dread. A teacher who only adjusts mechanics, adding another quiz, often gets dynamics nobody wanted. Reading the classroom through MDA surfaces the mismatch.

Teams and workplace systems

A compensation plan is a set of mechanics. The dynamics are the behavior it produces, which is frequently not the behavior leadership intended, because incentives, like game rules, generate emergent play. The aesthetics are how the work feels: motivating or cynical. Anyone who has watched a well-meaning bonus structure produce sandbagging has watched MDA dynamics in action.

The Elephant in the Room

Here is the thing nobody likes to say about MDA. It is a brilliant way to describe a game and a frustrating way to make one. The framework hands you a clean vocabulary for talking about what happened, but the moment you ask the question you actually care about, “what rules will produce the feeling I want,” it goes quiet. It tells you the layers are causally linked and then admits the causation runs through emergence, which it cannot give you a method to control.

This is not a knock on Hunicke, LeBlanc, and Zubek. It is the honest boundary of a descriptive framework. MDA does the half of the job that vocabulary can do: it lets a team see clearly. The other half, the part where you reliably engineer a target emotion, requires a model of human motivation that MDA never set out to provide. The aesthetics layer names eight emotions but says nothing about why a human craves any of them. Why does Challenge feel good? Why is Fellowship worth grinding for? Why does Discovery pull us forward when nothing forces it? Answer those, and you can finally build toward an aesthetic on purpose instead of by trial and error. That is precisely the question a motivation framework exists to answer, and it is where MDA hands the baton.

How to Apply MDA with the Octalysis Framework

MDA’s aesthetics tell you the destination. The Octalysis Framework tells you the road. Where MDA names eight kinds of fun as outcomes, Octalysis names eight Core Drives as the underlying human motivations that produce those outcomes. Stack them and the unfinished sentence in MDA finally gets a verb.

Map MDA’s aesthetics onto the Core Drives and the connection becomes obvious. The Challenge aesthetic is powered by Core Drive 2 (Development & Accomplishment), our drive to make progress and master skills. The Fellowship aesthetic runs on Core Drive 5 (Social Influence & Relatedness), our need to belong and connect. Expression draws on Core Drive 3 (Empowerment of Creativity & Feedback), the satisfaction of making something and seeing it respond. Discovery taps Core Drive 7 (Unpredictability & Curiosity), the pull of the unknown. Narrative and Fantasy lean on Core Drive 1 (Epic Meaning & Calling), the sense of being part of something larger than yourself.

This is the move that turns description into construction. MDA tells you “this game needs more Fellowship.” On its own, that is a wish. Octalysis tells you Fellowship is the felt surface of Core Drive 5 (Social Influence & Relatedness), and then hands you the specific techniques that activate it: group quests, mentorship, social treasure, shared goals that are easier together than alone. Now “add more Fellowship” becomes a build order. You are no longer hoping the aesthetic shows up. You are engineering the motivation that produces it.

The two frameworks also share MDA’s best instinct, the directional one. Octalysis is built for experience-first design, starting from the human motivation you want to activate and working toward the mechanics that activate it, which is MDA’s player-to-designer direction made into a method. Use MDA to name the eight kinds of fun you are targeting. Use Octalysis to find the Core Drives underneath them and the concrete techniques that switch those drives on. MDA gives you the target; Octalysis gives you the trigger.

Practical Steps to Run an MDA Pass

Here is how to actually use MDA on a project, whether it is a game, an app, or any system real people interact with.

1. Name the aesthetic first. Before touching a single rule, write down the specific feeling you want. Not “fun.” Pick from the eight: are you building for Challenge, Fellowship, Discovery, Expression, or some named blend? This is the target everything else serves.
2. Sketch the dynamics that would produce it. Ask what kind of behavior creates that feeling. Challenge needs escalating difficulty and clear feedback. Fellowship needs goals that are easier together than alone. Write the behavior you are hoping to see.
3. Build the minimum mechanics to trigger those dynamics. Now, and only now, design rules. Keep them lean. You are trying to spark the target behavior, not pile on features. Remember mechanics are the only layer you control directly.
4. Playtest and watch the dynamics you actually get. Put real people on it and observe without intervening. The behavior that emerges is the truth. The gap between the dynamics you wanted and the dynamics you got is your entire to-do list.
5. Tune the mechanics, never the players. When the dynamics are wrong, the fix is always at the mechanics layer. Adjust a number, add a constraint, remove a rule. Then test again. This loop, the one MDA was built around, is the work.
6. Add the motivation layer with Octalysis. When an aesthetic refuses to land, drop down to the Core Drive beneath it and ask whether you are actually activating the human motivation that produces the feeling. That is the causal step MDA leaves out and where most stuck designs come unstuck.

MDA Was the Beginning, Not the End

MDA earned its place by doing one thing better than anything before it: giving an entire field a shared way to talk about the invisible chain from rule to feeling. The three layers, the eight kinds of fun, and above all the insight that designer and player read the game in opposite directions, those ideas reshaped how games get made and how they get studied. Twenty years on, a designer who cannot fix a game can still locate the problem faster with MDA than without it.

But a vocabulary is not a method. MDA tells you where the problem lives and what feeling you are missing. It does not tell you why humans want that feeling or how to reliably manufacture it, and that missing causal layer is not a small thing. It is the difference between describing fun and designing it on purpose. The frameworks that came after, from DDE to Octalysis, are all attempts to finish the sentence MDA started. Name the aesthetic you want, then go find the Core Drive underneath it. That is where the real building begins.

Frequently Asked Questions

What does MDA stand for?

MDA stands for Mechanics, Dynamics, and Aesthetics. Mechanics are the rules and components a designer writes, Dynamics are the behavior those rules produce during play, and Aesthetics are the emotional responses the player experiences. The three form a causal chain from rule to feeling.

Who created the MDA framework?

Robin Hunicke, Marc LeBlanc, and Robert Zubek published the MDA framework in 2004 in the paper “MDA: A Formal Approach to Game Design and Game Research.” It grew out of Marc LeBlanc’s Game Design and Tuning Workshop, taught at the Game Developers Conference from 2001 to 2004.

What are the eight kinds of fun in MDA?

The eight aesthetics, often called the eight kinds of fun, are Sensation, Fantasy, Narrative, Challenge, Fellowship, Discovery, Expression, and Submission. They give designers a precise vocabulary for the type of enjoyment a game targets, replacing the vague word “fun.”

Why do designers and players experience MDA in opposite directions?

Designers build from the bottom up: they write Mechanics, which produce Dynamics, which create Aesthetics. Players experience the game from the top down: they feel the Aesthetics first, then notice the Dynamics, and only later understand the Mechanics. This directional asymmetry is MDA’s central insight about the craft of design.

What is the difference between mechanics and dynamics?

Mechanics are the explicit rules you write and can edit directly, such as cooldowns, card values, or spawn points. Dynamics are the emergent behavior those rules produce when a real player interacts with them, such as bluffing in poker or camping in a shooter. You control mechanics directly; dynamics you can only influence and observe.

What are the main criticisms of the MDA framework?

Critics note that the eight-aesthetic list is incomplete and somewhat arbitrary, that MDA offers no method for reliably getting from a desired feeling to the rules that produce it, and that it blurs the line between intended and actual player experience. The DDE framework (2017) and other successors were built to address these gaps.

How is MDA different from the Octalysis Framework?

MDA’s aesthetics describe the emotional experience a game targets, while Octalysis describes the human motivations that produce those experiences through its 8 Core Drives. MDA names the destination; Octalysis explains why people travel there and supplies the techniques to get them moving. The two frameworks stack rather than compete.

How do I actually use MDA in a project?

Start by naming the specific aesthetic you want, sketch the dynamics that would create it, then build the minimum mechanics to trigger those dynamics. Playtest to see the behavior you actually get, and tune the mechanics, never the players, until the target feeling lands. Pair it with a motivation framework like Octalysis to fill in the causal step MDA leaves out.

References

• Hunicke, R., LeBlanc, M., & Zubek, R. (2004). MDA: A Formal Approach to Game Design and Game Research. Proceedings of the AAAI Workshop on Challenges in Game AI, Vol. 4. AAAI Press.
• Hunicke, R., LeBlanc, M., & Zubek, R. (2004). MDA: A Formal Approach to Game Design and Game Research (full paper PDF).
• LeBlanc, M. (2004). Game Design and Tuning Workshop Materials, Game Developers Conference. Collected game design rants archived at 8kindsoffun.com.
• LeBlanc, M. (2004). Mechanics, Dynamics, Aesthetics: A Formal Approach to Game Design. Lecture, Northwestern University.
• Church, D. (1999). Formal Abstract Design Tools. Game Developer Magazine, August 1999.
• Walk, W., Görlich, D., & Barrett, M. (2017). Design, Dynamics, Experience (DDE): An Advancement of the MDA Framework for Game Design. In Korn & Lee (Eds.), Game Dynamics. Springer.
• Zubek, R. (2020). Elements of Game Design. MIT Press.
• Schell, J. (2008). The Art of Game Design: A Book of Lenses. Morgan Kaufmann.
• Salen, K., & Zimmerman, E. (2004). Rules of Play: Game Design Fundamentals. MIT Press.
• Bartle, R. (1996). Hearts, Clubs, Diamonds, Spades: Players Who Suit MUDs. Journal of MUD Research, 1(1).
• Csikszentmihalyi, M. (1990). Flow: The Psychology of Optimal Experience. Harper & Row.
• Aversa, D., et al. (2021). Redefining the MDA Framework — The Pursuit of a Game Design Ontology. Information, 12(10), 395.
• Chou, Y. (2015). Actionable Gamification: Beyond Points, Badges, and Leaderboards. Octalysis Media.

Related Reading

• The Octalysis Framework — the motivational engine beneath MDA’s aesthetics, mapping all 8 Core Drives that produce each kind of fun.
• Flow Theory — the engineering manual for MDA’s Challenge aesthetic, on keeping skill and difficulty in balance.
• Bartle’s Player Types — the player side of the equation: which aesthetics different players actually come for.
• Self-Determination Theory — the psychology of autonomy, competence, and relatedness underneath the eight kinds of fun.
• The Decoy Effect — how context shapes choice, a behavioral cousin to MDA’s emphasis on emergent player behavior.
• The Behavioral Framework Library — every psychological and design model in this series, mapped for builders.

WOULD YOU LIKE YU-KAI CHOU TO WORK WITH YOUR ORGANIZATION?