Lab or fieldwork walkthrough

Talk through a procedure and its failure points before doing it, so the first real attempt isn't also your first time noticing the missing step.

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: Useful once for a topic

When to use it

You are about to run a lab or field protocol for the first time.
You can recite steps but have not thought about where it fails.
You want a checklist that still requires understanding, not a substitute for it.

What to ask AI

Type something like: “Protocol: serial dilution for a bacterial count. I will narrate from memory. Note skipped steps. Then ask what-if failure branches one at a time. End with a compact bench checklist.”

Tell it not to reteach the whole method until after your narration.

You narrate first; interrogation second.

How it helps

Labs punish unexamined sequences: bad samples, unsafe moves, uninterpretable data.

A walkthrough is cognitive rehearsal with failure anticipation before your hands are busy.

AI cannot see your bench; local realities still need your attention.

How to run it

Narrate the procedure from memory without notes.
Receive notes on skipped steps only, not a full lecture.
Answer what-if branches one at a time.
Build a short checklist you understand line by line.
Perform, then forensic any failure afterward.

Example exchange

You

I’m going to narrate PCR setup from memory,no looking at the protocol. Catch steps I skip, then hit me with what-if branches one at a time.

You never mentioned changing tips between samples. What happens if you don’t?

You

You get cross-contamination between tubes, and then false positives that look like real amplification.

Good. Next branch only: what if the negative control amplifies?

Copyable prompt

Protocol: [ ].
I will narrate the procedure from memory. Note skipped steps and unsafe
shortcuts, don't reteach yet. Then interrogate me on failure points: ask "what
if" branches one at a time. Finally give a compact checklist I can take to the
bench, without replacing my need to understand why each line exists.

The Tell

You know this method has flipped when chat rehearsal felt fluent and the first real anomaly froze you.

Fluency in narration is not readiness at the bench.

Demand failure branches, then practise the awkward ones aloud.

Principle evidence

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

The underlying learning idea is rated moderate. Mental walkthrough of procedures before performance relates to mental practice / cognitive rehearsal and pre-lab briefing in science education. Principle moderate; AI failure-point lists may miss local lab realities, speculative.

AI delivery evidence

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

The learning principle may be moderate, but that does not prove an AI session delivers it well. Treat AI delivery as speculative unless a study of tutoring with this method is named, and none is claimed here.