What the twin knows, you know.
Every learner gets a private model of their own understanding — built only from what they did: reads, quiz answers, teach-backs, tutor questions. It distinguishes assisted from unassisted success, flags misconceptions as suspected until evidence confirms them, and schedules the next week around the gaps it can prove.
AI student: Everything on this page is mine to keep — and yours to read, correct, or dispute. Try the controls in the card; they are the real ones.
Already a learner? Your real twin is at your twin page after you log in.
AI student: An illustrative learner, fully expanded. Nothing here is live — the real twin asks before it remembers.
Learning Twin — Maya, Biology
Everything the twin stores about one learner, in one view.
Signals — what the twin logged, and what each kind counts as
- Osmosis vs. diffusionMon
Read the lesson — Exposure — never moves the score on its own.
- Osmosis vs. diffusionTue quiz
Solved without help — The strongest kind: unassisted, first try.
- Osmosis vs. diffusionFri teach-back
Held three days later — Retention — you re-derived it unaided, days later.
- Electron transport chainWed quiz
Solved with hints — Counted, but never confused with mastery.
- Linked gene inheritanceWed quiz
Solved with hints — Counted, but never confused with mastery.
Misconceptions — suspected first, confirmed only by evidence
Said twice in a teach-back on Wed. Suspected, not confirmed — a later wrong answer on the concept would confirm it; one clean unassisted answer would clear it.
Your say — separate from the twin's score
Osmosis vs. diffusion
System score: 0.82 (evidence-backed). How sure are you, yourself?
Think the twin has you wrong? Disputing never changes the score — it puts a mark on the concept your teacher can see.
This card is the whole model. Nothing about you lives anywhere else.