Methods · Metacognition & self-regulation

Study-process audit

Describe how you studied, honestly, and have the method itself critiqued rather than the content.

When to use this: At the end of a session
Learning impact: Strong change
Learning evidence: Credible with caveats
Where you are with the topic: Checking readiness
How often to use it: Worth repeating often

When to use it

A study session just ended and you want the *method* critiqued, not more help with the content.
You planned to practise problems and somehow spent forty minutes reorganising notes again.
Results are flat despite long hours, and you suspect how you study is the leak.

What to ask AI

Describe exactly what you did, including the drift: “Here's what I did in my last two-hour session… Don't help me with the content. Critique the METHOD.”

Ask how much was retrieval, how much was reading, and how much was neither, and to be blunt about the neither.

Ask for exactly one change next session, not a new system.

How it helps

You get endless feedback on content and almost none on method. Marks tell you what you knew; nothing tells you how you came to know it, or failed to.

A weak technique for six hours loses to a strong technique for two, and both people conclude something about ability.

What you intended and what you did diverge toward the comfortable: practice became re-reading; recall became glancing at the page. Nobody watches, so nobody notices.

The audit makes the drift visible. Then you change one thing and check whether it survived a tired Tuesday.

How to run it

Describe the session concretely, order, duration, what was open in front of you, not the plan you meant to follow.
Include the drift: the reread, the switch when it got hard, the tidying. That is the data.
Ask for the method to be critiqued, not the content. Say so explicitly or content help will take over.
Ask what the session was actually good for. Most methods fit some situation; naming it stops a blanket “that was useless.”
Ask for one change only. Re-audit in a week to see whether it stuck.

Example exchange

You

Last night I did two hours of chemistry: twenty-five minutes rewriting notes, forty re-reading with highlights, thirty of 'practice' with solutions open, twenty-five organising folders. Can you critique the method, not the content?

Most of that trains recognition and tidy admin. Retrieval and closed-book problems barely happened. While you were doing it, how much felt productive?

You

Almost all of it. The folder bit felt like a break that still counted as studying.

That feeling is the trap. One change next time: first forty minutes, three problems closed-book, solutions face-down. Everything else waits. Re-audit only after that session.

Copyable prompt

Here's exactly what I did in my last two-hour session: [describe honestly,
including the drift]. Don't help me with the content. Critique the METHOD: what
was I actually practising, what did this session train, and what did it not
touch? Then give me exactly one change for next session, not a system.

The Tell

If three audits find the same problems and nothing in your next session changes, the audit has become a ritual.

Pick the single smallest change and do only that, then re-audit.

Principle evidence

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

Moderate: auditing study process is metacognitive evaluation of strategy in self-regulated learning. The leverage is that method often explains results better than “ability,” yet almost never gets feedback.

AI delivery evidence

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

AI advice about study methods can be generic or trendy. Insist on critique of your described session, one concrete change, and a re-check. Delivery quality is speculative; honesty in your description is not.