About This Method

Instrument Deep Learning Framework

A structured methodology for deeply learning any electronic instrument — synthesizers, drum machines, effects units, or modular systems. Designed for people who buy gear they love but don't go deep enough to use it fluently in compositions.

See also: adhd-design.md — ADHD-specific design rationale that shapes every session.

The Problem

You buy an instrument. You love the sounds. You noodle. You never truly learn it. It sits underused while you reach for the familiar. This framework fixes that.

Core Principles

1. Isolation Before Integration

Learn each subsystem in isolation before combining. You can't debug a patch if you don't know what each component does independently.

2. Basic Patch as Home Base

Every instrument needs a "basic patch" — a minimal, known-good starting state that isolates the thing you're studying. Always return here before exploring a new concept.

3. Micro-Sessions with Tangible Output

Sessions are 15-30 minutes. Every session ends with something concrete: a documented patch, a technique you can name, a recording snippet, or a composition element. No noodling sessions.

4. Progressive Complexity

Follow the signal flow. Start with oscillators (sound generation), move to filters (sound shaping), then modulation (movement), then effects (space), then sequencer (pattern), then integration (composition). Each layer builds on the last.

5. Spaced Repetition Through Warm-Ups

Each session begins with a 2-3 minute warm-up that reinforces the previous session's concept. This prevents the "I learned this last week but forgot" problem.

6. Document Everything

Patches are ephemeral if undocumented. Every patch worth saving gets a full parameter dump. Every technique gets a name and a description. This builds your personal reference library.

7. Composition-Driven Goals

The endgame is always "use this in a song." Sessions progress from isolated exercises to compositional applications. The final modules are always about integration with your DAW and existing workflow.

Framework Structure

For Each Instrument

instruments/<name>/
├── overview.md          # What this instrument is, architecture, signal flow
├── signal-flow.md       # Visual/textual signal flow diagram
├── parameters.md        # Complete parameter reference (from manual)
├── basic-patch.md       # The "home base" patch with all parameter values
└── modules.md           # Learning module sequence with dependencies

sessions/<name>/
├── 01-foundations-*.md   # Module 1 sessions
├── 02-oscillators-*.md   # Module 2 sessions
├── ...
└── XX-integration-*.md   # Final integration module

patches/<name>/
├── basic-patch.md        # The reference basic patch
├── exercise-*.md         # Patches created during exercises
└── song-*.md             # Patches used in actual compositions

Module Taxonomy

Every instrument maps to this module sequence (skip what doesn't apply):

#ModuleFocusPrerequisite
1FoundationsBasic operation, navigation, saving/loadingNone
2Sound SourcesOscillators, noise, samples, external input1
3Sound ShapingFilters, waveshaping, distortion2
4EnvelopesADSR, amplitude, time-based shaping2
5ModulationLFOs, mod matrix, velocity, aftertouch3, 4
6EffectsDelay, reverb, chorus, distortion3
7Sequencer/ArpStep sequencer, arpeggiator, pattern5
8Sound DesignCombining everything, recipe-based patches5, 6
9PerformanceLive playing techniques, expression, MIDI8
10IntegrationDAW workflow, recording, using in songs8, 9

Session Template

Every session follows this structure:

# Session XX: [Topic]
Module: [module name] | Duration: [15-30] min | Prerequisite: [session #]

## Objective
One sentence: what you will be able to do after this session.

## Warm-Up (2-3 min)
Quick exercise reinforcing the previous session's concept.

## Setup
Starting patch state (basic patch or specified modifications).

## Exercises
Numbered steps with specific parameter values and expected results.

## Exploration Prompts
"Try this..." suggestions for going beyond the exercise.

## Session Output
What to save/document before ending:
- [ ] Patch saved? (name and location)
- [ ] Technique documented?
- [ ] Obsidian daily note updated?

## Next Session Preview
One sentence about what's coming next and why it builds on this.

Adding a New Instrument

  1. Create instruments/<name>/overview.md — read the manual, document architecture
  2. Create instruments/<name>/signal-flow.md — trace the audio path
  3. Create instruments/<name>/basic-patch.md — define your home base
  4. Create instruments/<name>/modules.md — map the module taxonomy to this instrument's features
  5. Create sessions following the module sequence
  6. Add Obsidian templates if needed

Progress Tracking

Progress is tracked in an Obsidian vault through:

  • Daily notes: Session log entries with tags
  • Instrument progress notes: Per-instrument tracker with module completion
  • Patch library: Linked notes for documented patches

See obsidian/ for templates.

Philosophy

This framework exists because deep learning of an instrument requires:

  • Structure to prevent aimless noodling
  • Accountability to maintain momentum across sessions
  • Documentation to compound knowledge
  • Composition focus to ensure learning serves music-making

The goal is never "know every parameter." The goal is "reach for this instrument confidently when making music, and know how to get the sound in your head out of it."


ADHD-Aware Learning Design

This document captures the specific design decisions that make this framework work for learners with ADHD or attention challenges. Every session, every structure, every output requirement is shaped by these principles.

The Core Problem

ADHD doesn't mean you can't focus — it means you can't choose what to focus on. Hyperfocus on a new synth patch is easy. Returning to methodical practice on Tuesday when Monday was exciting? That's the hard part.

This framework is designed for the hard part.

Design Principles

1. Zero Activation Energy

The "wall of awful" — the emotional barrier to starting — is the #1 enemy of consistent practice. Every session is designed to minimize startup friction:

  • No decisions at the start. The session tells you exactly what patch to load, what knob to turn first. Decision fatigue kills momentum before you play a single note.
  • 2-minute warm-ups. You're making sound within 120 seconds. Not reading, not setting up, not deciding — playing.
  • "Open the doc and do step 1" — that's the only commitment. Momentum handles the rest.

2. Dopamine by Design

ADHD brains need reward signals. Abstract learning ("now you understand FM synthesis") doesn't register. Tangible output does:

  • Every session produces a named artifact — a saved patch, a documented technique, a recording. You can point to it and say "I made that."
  • Patches have names you choose. Naming things creates ownership and emotional connection.
  • The checklist at the end gives completion dopamine. Checking boxes is satisfying. Use it.

3. Time-Boxing is Non-Negotiable

  • 15-30 minutes maximum. Not because the content requires it, but because consistency over 10 weeks beats intensity over 2 weeks.
  • Sessions have a hard stop. "Exploration prompts" are optional extensions, not requirements. If you hit 30 minutes, stop. Save. Log. Done.
  • It's okay to stop mid-session. Note where you stopped. Pick up there next time. Partial progress > no progress.

4. Structure Prevents Drift

The #1 failure mode for ADHD learners with synthesizers: "I'll just noodle for a bit" → 2 hours of unfocused sound exploration → nothing documented → no sense of progress → guilt → avoidance.

  • Sessions have a single objective stated as one sentence. If you can't say what you're learning in one sentence, the session is badly designed.
  • Exercises have specific parameter values. "Try changing the filter" is too vague. "Set LPF Frequency to 50, Resonance to 82" is actionable.
  • Exploration prompts are bounded. "Try 3 different waveshapes" not "explore all the waveshapes."

5. Warm-Ups Bridge the Forgetting Gap

ADHD working memory is volatile. What you learned Thursday is fog by Monday. Warm-ups solve this:

  • Every session starts by doing something from the previous session — not reading about it, doing it.
  • Warm-ups are 2-3 minutes, not quizzes. Play the patch you made last time. Turn the knob you learned about. Physical action triggers recall.
  • If the warm-up feels unfamiliar, repeat the previous session instead of pushing forward. No shame, only reinforcement.

6. Progress Must Be Visible

ADHD brains discount invisible progress. "I'm 60% through a learning program" means nothing unless you can see it.

  • Module completion in Obsidian — checkboxes you can see filling up
  • Patch library growing — tangible evidence of accumulated skill
  • Session logs in daily notes — a trail of effort that compounds

7. Multiple Entry Points

Some days you have 15 minutes of focused energy. Some days you have 5 minutes before context-switching. Some days you have an unexpected hour of hyperfocus.

  • Sessions are designed for 15-30 min but can be split across days
  • Exploration prompts are there for hyperfocus days — structured rabbit holes
  • Quick-reference cards (patch values, signal flow) are available for 5-minute refreshers without starting a full session

8. Forgiveness is Built In

  • Skipping days/weeks is expected, not a failure. The curriculum doesn't have dates, only sequence.
  • Repeating sessions is encouraged. Some concepts need 3 passes. The basic patch warm-up makes re-entry low-friction.
  • "Good enough" patches count. You don't need the perfect FM bell — you need the experience of making one.

Session Structure (ADHD-Optimized)

┌─────────────────────────────────┐
│  OBJECTIVE (1 sentence)         │  ← Know exactly why you're here
├─────────────────────────────────┤
│  WARM-UP (2-3 min)              │  ← Immediate action, no decisions
│  Play something from last time  │
├─────────────────────────────────┤
│  SETUP (1 min)                  │  ← Exact patch state specified
│  Load patch, verify             │
├─────────────────────────────────┤
│  EXERCISES (10-20 min)          │  ← Numbered steps with values
│  Step 1: Do this specific thing │     Each step has an audible result
│  Step 2: Change this value      │     You're always making sound
│  Step 3: ...                    │
├─────────────────────────────────┤
│  EXPLORATION (optional, 5 min)  │  ← For hyperfocus days only
│  Bounded suggestions            │
├─────────────────────────────────┤
│  OUTPUT CHECKLIST               │  ← Dopamine! Check the boxes
│  □ Patch saved                  │
│  □ Technique noted              │
│  □ Obsidian logged              │
└─────────────────────────────────┘

Anti-Patterns to Avoid

Anti-PatternWhy It FailsWhat To Do Instead
"Explore freely"No structure → drift → no outputGive 3 specific things to try
Long setupActivation energy kills sessionSetup is "load patch, play note"
Reading-heavy sessionsADHD brains learn by doingImmediate hands-on within 2 min
Perfectionism gates"Master this before moving on" → stalling"Good enough" = done
Calendar-based schedulesMissed dates → guilt spiralSequence-based, no dates
No visible progressDiscount invisible learningObsidian checkboxes + patch library

For Session Authors

When writing new sessions for this framework:

  1. Can someone start making sound within 2 minutes of opening this document?
  2. Does every exercise step have a specific, audible result?
  3. Is the objective one sentence?
  4. Does the output checklist have 3-4 items maximum?
  5. Could someone stop at any exercise step and still have gained something?
  6. Is the warm-up a physical action, not a reading task?

If any answer is "no," redesign the session.

Run It Yourself

This is a demo showing the Evolver learning system with synthetic practice data. To use it with your own Evolver and Obsidian vault:

  1. Clone the repository: git clone https://github.com/lovettbarron/evolver.git
  2. Install dependencies: npm install
  3. Create evolver.config.json with your vault path
  4. Run npm run dev