experience-design

What this skill does

# Experience Design

**Purpose:** Tools for designing, diagnosing, and improving the player *experience* — not the feature set. Games are systems that generate experiences; this skill helps you build better generators.

**Influences:** Frameworks here draw on experience engineering theory, cognitive engagement research, and systematic game balance methodology.

---

## When to Activate

Use this skill when:
- Designing or evaluating a core gameplay loop
- Something "works" mechanically but doesn't feel engaging
- Planning the emotional arc of a level, session, or campaign
- Evaluating whether a mechanic creates meaningful choices
- A feature is technically complete but players aren't having fun
- Deciding what to cut vs. what to keep

---

## Core Framework: The Experience Triangle

Every moment of gameplay is the product of three forces:

```
        Mechanics
       (what players do)
          /    \
         /      \
        /        \
   Fiction ——— Feedback
 (what it means)  (what players perceive)
```

- **Mechanics** — The systems and rules. What inputs are available? What outcomes are possible?
- **Fiction** — The narrative and thematic wrapper. What does this action *mean* in the game world?
- **Feedback** — What the player sees, hears, and feels in response. How does the system communicate?

**Diagnostic:** When something feels wrong, identify which vertex is weak:

| Symptom | Weak Vertex | Fix Direction |
|---------|-------------|---------------|
| "I don't know what to do" | Mechanics (unclear rules) | Simplify options, add tutorials |
| "I don't care" | Fiction (no meaning) | Connect to stakes, add narrative weight |
| "Did that work?" | Feedback (no response) | Add visual/audio/haptic confirmation |
| "It's boring" | Mechanics + Fiction gap | Close the gap — make actions feel consequential |
| "It's confusing" | All three misaligned | Strip back and realign around one clear experience |

---

## The Engagement Loop

Every game runs on loops. The tightest one is the core engagement loop:

```
Action → Feedback → Evaluation → Decision → Action ...
```

1. **Action** — Player does something (the mechanic)
2. **Feedback** — System responds (perceivable, timely, proportional)
3. **Evaluation** — Player assesses the outcome (cognitive processing)
4. **Decision** — Player chooses what to do next (meaningful choice)

### Loop Health Check

- [ ] Can a player complete the loop in < 10 seconds? (Core loop should be fast)
- [ ] Is the feedback immediate and perceivable?
- [ ] Does evaluation involve genuine uncertainty? (If outcome is always obvious, the loop is dead)
- [ ] Does the decision feel like a real choice? (Not "one right answer")
- [ ] Does the loop build toward a larger loop? (Session loop, progression loop)

### Nested Loop Structure

```
Core loop:     move → shoot → evaluate → reposition     (seconds)
Session loop:  mission → rewards → loadout → next mission  (minutes)
Meta loop:     campaign → unlock → new content → campaign   (hours/days)
```

Each loop should be satisfying on its own while feeding into the next.

---

## Emotion Layering

The best moments layer multiple emotions simultaneously. Design for this:

1. **Name the primary emotion** — not "fun" (too vague), but *mastery satisfaction*, *discovery delight*, *social triumph*, *creative pride*
2. **Add a contrasting undercurrent** — tension + anticipation, relief + pride, curiosity + apprehension
3. **Use fiction to bridge them** — narrative context turns mechanical outcomes into emotional ones
4. **Leave room for emergence** — don't over-specify; let the system surprise

**Anti-pattern:** Designing for "fun." Fun is an umbrella over dozens of specific emotions that require different design approaches.

---

## The Elegance Test

For any mechanic or feature, evaluate:

```
Elegance = Emotional Payoff / (Player Attention Cost + Development Cost)
```

| Factor | Question |
|--------|----------|
| **Emotional payoff** | What does the player feel? How intensely? How often? |
| **Attention cost** | How much must the player learn, remember, and track? |
| **Development cost** | How much effort to build, tune, and maintain? |

- **High elegance:** Simple to learn, cheap to build, produces rich emotional variety
- **Low elegance:** Complex to understand, expensive to build, produces tepid engagement

**Decision rule:** If a feature scores low on elegance, either simplify it or cut it. Complexity is a budget — spend it on what matters.

---

## Meaningful Choice Evaluation

A choice is meaningful when:

1. **The player understands the options** (clarity)
2. **The options have different consequences** (differentiation)
3. **The consequences matter** (stakes)
4. **No option is obviously best** (tension)
5. **The player has enough information to reason but not to solve** (uncertainty)

### Choice Diagnostic

| Question | If "No" |
|----------|---------|
| Can the player understand what each option does? | Clarity problem — simplify or explain |
| Do the options lead to genuinely different outcomes? | False choice — collapse or differentiate |
| Does the outcome affect something the player cares about? | Stakes problem — raise consequences |
| Is there a dominant option? | Balance problem — see **game-balance** |
| Is the outcome fully predictable? | Determinism problem — add uncertainty or hidden information |

---

## Pacing

Pacing is cognitive load management across time.

### The Tension Curve

```
Intensity
    |     /\        /\  /\
    |    /  \  /\  /  \/  \
    |   /    \/  \/        \___
    |  /
    | /
    +---------------------------→ Time
     Intro  Rising  Climax  Rest
```

**Principles:**
- **Peaks need valleys.** High-intensity moments are meaningless without contrast
- **Valleys are active.** Rest periods aren't empty — they're exploration, narrative, planning, customization
- **Escalate across sessions.** Each session's peak should be slightly higher than the last
- **Respect encoding time.** After introducing a new mechanic or concept, give the player space to practice before adding more

### Pacing Checklist

- [ ] Does the first 5 minutes teach through *doing*, not *reading*?
- [ ] Is there a clear rhythm of tension and release?
- [ ] After each new concept, is there practice time before the next one?
- [ ] Do difficulty spikes coincide with new tools/abilities (not just harder enemies)?
- [ ] Is there a moment of "earned rest" after each major challenge?

---

## "Why Isn't This Fun?" Diagnostic

When a mechanic works correctly but doesn't engage, run through in order:

1. **Is the core loop running?** (Action → Feedback → Evaluation → Decision)
   - If any step is broken, fix that first
2. **Is there a meaningful choice?** Run the Choice Diagnostic above
3. **Check the Experience Triangle** — which vertex is weak?
4. **Check pacing** — is this moment in a valley that's gone on too long? Or a peak with no buildup?
5. **Check elegance** — is the attention cost too high for the payoff?
6. **Check emotion targets** — what *specific* emotion should this produce? Is the mechanic actually capable of producing it?
7. **Check context** — is the fiction supporting or undermining the mechanic?

**If all seven check out and it still isn't fun:** The mechanic may simply not be interesting. Consider cutting it and redirecting the complexity budget elsewhere.

---

## Narrative Integration

Three approaches to narrative, each with different design implications:

| Mode | Control | Replayability | Design Challenge |
|------|---------|---------------|------------------|
| **Scripted** | High (author-controlled) | Low | Don't let narrative contradict mechanics |
| **Environmental** | Medium (player-discovered) | Medium | Reward exploration without requiring it |
| **Emergent** | Low (system-generated) | High | Build systems that produce story-worthy events |

**The dissonance test:** If the story says one thing and the mechanics reward another, players will follow the mecha