Persistent persona continuity

Work with the same expert or character across many sessions so prior context deepens, without letting continuity become uncritical trust.

When to use this: Before you start studying
Learning impact: Unlocks better later work
Learning evidence: Thin or mixed evidence
Where you are with the topic: Building understanding
How often to use it: Build and keep as a habit

When to use it

You restart every chat as if the tutor never met you.
You want continuity with one coach, examiner, or character across sessions.
Setup overhead is eating practice time.

What to ask AI

Type something like: “Persona card: sceptical methods tutor who knows I confuse p-values with effect size. Persist that card. On each session, ask what we last left unresolved before new content.”

Paste the card each time if the tool forgets.

How it helps

Persistent persona continuity is mostly an AI affordance: cumulative context so practice deepens.

It is an enabler for other methods, not a learning principle by itself.

Protect against false memory of what “your tutor” believes about you, you still verify.

How to run it

Write a short persona card: role, stance, known gaps.
Paste it at the start of each session.
Require a recap of unresolved items before new work.
Update the card when a gap closes or a new one appears.
Do not treat persona memory as a gradebook of truth.

Example exchange

You

Persona card: viva coach. Last time I hedged on measurement. Continue from that card,don’t reset.

Before new content: last unresolved item was field potentials versus single-unit claims. Thirty-second answer now.

You

Field potentials average activity across many cells, so they can’t isolate what one neuron is doing.

Clearer. Updating the card: measurement hedge improved; next weak node is experimental controls. Ready for one control question?

Copyable prompt

You are [PERSONA]. Card: [ ]. Last session I committed to: [ ].
Stay in role. Open by checking that commitment. Teach with contingent hints
(A2). At end, propose 3 bullets to add to the card. I will edit before saving.
Do not claim degrees or sources you can't support; say "simulate only."

The Tell

You know this method has flipped when the persona “remembers” progress you did not earn, and you feel prepared because the chat is warm.

Continuity is not competence.

Keep the card honest: unresolved items stay unresolved until you demonstrate them.

Principle evidence

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

The underlying learning idea is rated weak. Persistent persona continuity is mostly an AI affordance (memory/state). Educational benefit is hypothesized via relationship continuity and cumulative context, thin direct evidence. Principle weak/speculative; treat as enabler for other methods.

AI delivery evidence

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

The learning principle may be weak, 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.