Methods · Support & cognitive load

A13

Chunking a procedure

Break a long procedure into named stages you can recite from memory, so mid-task panic becomes "I'm in stage 3" instead of "I've forgotten everything.".

When to use this
First time through the material
Learning impact
Reliable gain
Learning evidence
Credible with caveats
Where you are with the topic
Building understanding
How often to use it
Worth repeating often
What AI is doingCreate practice, examples, or schedules

When to use it

  • You are meeting this idea for the first time and need a way in.
  • The reading feels dense and you want a first map before the details.

What to ask AI

Type something like: “Here is a procedure I need to perform unaided. Reorganise it into 3-6 named”.

Ask it to generate material for you to attempt, problems, cues, or a schedule, not finished answers.

Keep the hard part on your side: you retrieve, attempt, explain, or decide before you ask for help.

How it helps

Long procedures overwhelm for a boring reason: working memory cannot hold twelve unordered steps while also executing them.

Experts look fluent because they don't hold twelve steps, they hold four named chunks, each of which expands into substeps on demand.

Novices try to memorise the flat list.

Under pressure the list shuffles.

Chunking is the repair: impose a hierarchy, name the stages, rehearse the names until they retrieve the substeps.

AI helps you propose a clean stage structure and then quizzes the skeleton before you practise the full performance.

This is procedural sister to Dependency mapping dependency mapping (conceptual trees) and cousin to Procedure narration under constraint procedure narration (embodied skills).

A13 is the cognitive scaffold for any multi-step method you must run unaided.

How to run it

  1. Paste or describe the full procedure as you currently know it (messy is fine).
  2. Ask for a chunking proposal into 3-6 named stages, and reject more than six.
  3. Next: edit the names until they are words you will remember under stress.
  4. Next: recite the spine until perfect. Only then expand substeps from memory.
  5. Run a dry procedure saying stage names aloud as you go.
  6. Next: stress test: scramble. AI names a mid-stage, you produce its substeps and what must already be true from prior stages.

Example exchange

You

Here's how I currently do a titration, it's a long scramble. Reorganise it into a few named stages I can recite, then quiz me on stage order.

AI

Try these stages: Prepare, Titrate, Detect endpoint, Calculate. Recite them back, then tell me what Detect endpoint is for in one line.

You

Prepare, Titrate, Detect endpoint, Calculate. Detect endpoint means notice the colour change so I know when to stop adding.

AI

Good. What's the most common failure in that stage?

Copyable prompt

Here is a procedure I need to perform unaided. Reorganise it into 3-6 named
stages. For each stage: purpose in one line, substeps, a check that the stage
is complete, and the most common failure at that stage. Prefer memorable stage
names over technical ones.
PROCEDURE: [ ]

The Tell

Here is how you know this method has flipped: mid-procedure you can only continue by scrolling a tutorial. You have steps, not chunks.

In that moment the AI (or the schedule, or the story) did the thinking, and you only recognised a finished product.

Tighten the prompt for chunking a procedure so you attempt, decide, or retrieve before anything is handed to you.

Principle evidence

Strength of the underlying learning idea, not a claim about AI products.

The underlying learning idea is rated moderate. Chunking procedures reduces working-memory load (Miller’s chunking as background; CLT on element interactivity; motor/procedural training on part-task practice). Reciting named stages adds a light retrieval demand.

AI delivery evidence

Whether an AI tutor delivers this method well is a separate question.

Principle moderate; AI stage decomposition quality speculative.

Related methods