Goal: re-collect a revised fictional set without rewriting the first record. Proposed code: AC9MFST01. Prepare: Card 138, original set and documented new trial with one triangle replaced by a square.
- Retrieve · 3 min. Rebuild yesterday's unchanged display and describe one comparison from its actual cards.
- Model · 4 min. Announce new trial and swap one card explicitly; keep old display intact for comparison, not as a mistaken past result.
- Guide · 5 min. Child sorts new six-card set, checks each card is once and aligns categories.
- Independent compare · 7 min. Child identifies whether groups now have more/fewer/same, using matching/counting and a named dataset.
- Source check · 4 min. Ask whether the first fictional visitor changed their mind. No; this is a different teacher-made set. Do not claim a real class became evenly split.
- Exit · 2 min. Child points to the exact swapped category and the changed comparison. If old/new blend, use two labelled mats tomorrow.