Cognitive Engineering & Self-Mastery — Part 2B: Working-Memory Engineering & Knowledge Compression

Track: Slot Limits · Chunking · Compression Pipelines · Retrieval Maps · Output Artifacts

The mind performs best when it carries almost nothing. Capacity comes from structure, not strain.

1) Engineer the Slot Limit (Design for 3)

• Active Slot Ceiling: 3. If you show your brain more, it silently drops something important.
• Atomic Cards: each slot holds one verb + object + constraint (Draft intro (≤150w)).
• Icebox: every extra idea goes to a dated inbox; never crowd the 3.

2) Chunking: Expand Capacity Without Load

Group small items into a meaningful unit. Chunking is compression with a name.

• Principle: if items are used together, store them together.
• Test: can you recall the chunk title when stressed? If not, compress further.

CHUNK NAME: "Deep Block Start"
Parts → timer, token, two wins, doc open, phone out
Compression → "Timer·Token·TwoWins·Doc·Phone"
  

3) Compression Ladder (Advanced)

• 50 → 25 → 10 → 5 → 3 (headlines → thesis → mantra → trigram).
• Loss Aware: never delete safety-critical details; move them to checklist tier.

Topic: Attentional Shift
50: Shifting drains energy; schedule it; warm start; keep ramp ≤60s.
25: Change by plan, not impulse; script the first line; measure ramp time.
10: Plan shift → write next line → start timer.
5: Plan → Next Line → Start
3 (trigram): Plan•Line•Start
  

4) Retrieval Surface Area (Make Ideas Easy to Grab)

• Three Handles/Idea: a mantra, a visual, a first step.
• Mantra: ≤5 words; Visual: simple sketch; First step: 20–120 sec action.

Idea: Research Synthesis
Mantra → "Compare → Compress → Conclude"
Visual → 3 columns (S1/S2/S3) → arrow → 25-word thesis
First step → Paste quotes into 3 columns
  

5) Note Archetypes (Don’t Mix Types)

• Capture Note: raw, timestamped, source.
• Concept Note: single idea with 50→5 ladder + links.
• Playbook: checklist for action with “Done =”.
• Evidence Note: observations, scores, decisions; used for learning loops.

6) Graph Linking (Atomic → Map → Playbook)

• Each Concept Note links to 2 neighbours (why/where used).
• Maps summarise 5–9 Concept Notes; Playbooks reference Maps.

[Concept: Two-Context Rule] → [Map: Focus Systems] → [Playbook: Deep Work Start]
  

7) Error-Correcting Summaries (ECS)

Write the summary you would use to catch your future self’s mistake.

1. State the core claim in ≤25 words.
2. List 3 common errors.
3. Write a one-line fix for each error.

Claim: "Interleaving improves transfer."
Errors: (1) too many topics (2) no retrieval (3) random order
Fix: (1) cap to 3 (2) add 3 recalls (3) rotate A-B-A with goal
  

8) The 3-Tier Artifact Stack

• Tier 1 — Mantra: ≤5 words (carry in head).
• Tier 2 — Checklist: 5–9 steps (printable).
• Tier 3 — Map/Playbook: 1 screen; links to notes.

9) Compression KPIs

• Ramp-Up Latency ≤ 60s
• Steps/Task ≤ 9
• Recall (25w thesis) ≥ 90% on D+7
• Search Time for Playbook ≤ 30s

Templates (Copy/Paste)

Concept Note

# Concept: __________________________
50: ________________________________________________
25: ____________________________________
10: ______________________
5: _______
3: ___
Links: [Upstream] [Downstream]
Examples: 1) ______ 2) ______ 3) ______
  

Playbook

PLAYBOOK: __________________________
Goal: ______________________________
Steps (≤9):
1) ___________________
2) ___________________
3) ___________________
Done = _____________________________
Checks: time ≤ __; errors avoided: ____
  

Evidence Note

DATE/TIME: ________
Context: Focus / Recovery / Mixed
What I tried: ______________________
Outcome (1–5): __
Latency (s): __
Error seen: _______________
Fix next time: ____________
  

Free Execution Prompt — Working-Memory & Compression Architect (10-Year Proof)

You are my Working-Memory & Knowledge Compression Architect (Part 2B).
Goal: reduce cognitive load by enforcing a 3-slot system and building a compression pipeline (mantra → checklist → map).

ASK ME FIRST (inputs):
1) Today’s project and deadline
2) My typical ramp-up latency and tab count
3) A recent task that felt heavy
4) Tools I already use for notes
5) Where I will keep Playbooks (location)

DO THIS (steps):
1) Create three Atomic Cards (verb + object + constraint) for today; everything else to an Icebox list.
2) Build a Compression Ladder (50→25→10→5→3) for the project’s core idea.
3) Draft a Tier-2 Checklist (≤9 steps) with “Done = …” and timing checks.
4) Draw a one-screen Map (as text) linking 5–9 Concept Notes I’ll need; include two neighbours per note.
5) Write an Error-Correcting Summary (claim, 3 common errors, fixes).
6) Produce KPIs and a 7-day review plan (latency ≤60s; steps ≤9; recall on D+7 ≥90%; Playbook search ≤30s).

OUTPUT / ARTIFACTS:
- 3 Atomic Cards + Icebox
- Compression Ladder
- Tier-2 Checklist + Done definition
- Map (text links) + two neighbours/note
- ECS (errors + fixes)
- KPI set + 7-day review plan

EVIDENCE GRADING:
- High: start ≤60s; zero hunting for steps; recall 25-word thesis on D+7; consistent outputs from same Playbook.
- Moderate: start ≤90s; minor searching; recall ≥70%.
- Low: above thresholds → compress further; split steps; add mantras and visual handle.

LINK-FORWARD:
Advance to Part 2C (Knowledge Integration & Transfer — turning artifacts into cross-domain skill).
    

FAQ (Part 2B)

Next: Part 2C — Knowledge Integration & Transfer.

Apply It Now (5 minutes)

1. One action: What will you do in 5 minutes that reflects this essay? (write 1 sentence)
2. When & where: If it’s [time] at [place], I will [action].
3. Proof: Who will you show or tell? (name 1 person)

You are my Micro-Action Coach. Based on this essay’s theme, ask me:
1) My 5-minute action,
2) Exact time/place,
3) A friction check (what could stop me? give a tiny fix),
4) A 3-question nightly reflection.
Then generate a 3-day plan and a one-line identity cue I can repeat.

🧠 AI Processing Reality… Commit now, then come back tomorrow and log what changed.