# Foundation / Prep Science · ten 25-minute scripts, Days 71–80

First read [safe optional protocol and full original fictional sources](MATERIALS.md). Each lesson totals **3 + 4 + 5 + 7 + 4 + 2 = 25 minutes**. The [three same-target routes per day](LEARNER-CARDS.md) are flexible access choices, not fixed learner types; [worked swaps](OPTIONAL-PRACTICE.md) are extra examples. **Quiet paper reasoning is always available.** A live sound/action requires school approval and each child's comfort; never force hearing/touch. When no genuine action occurs, mark **REAL OBSERVATION NOT AVAILABLE**, and do not credit actual material response or AC9SFI02. An attributed neutral adult description stays **ADULT REPORT**, not the child's direct sense. Keep [Days 75/80 held checks](ASSESSMENT.md) unseen until first independent response; [keys](TEACHER-KEY.md) are separate. The [family bridges](FAMILY-OPTIONAL.md) are invitations, not homework.

## Week 15 · Day 71 — What can a tap arrow tell us?

**Goal:** separate a paper action symbol from an actual observed material response; pose a safe source question. **Proposed codes:** AC9SFH01, AC9SFI01, AC9SFI05. **Prepare:** complete Model R, [paper-model](paper-model.svg), [quiet-source](quiet-source.svg), Card 71.

1. **Quiet choice · 3 min.** State no tap will occur today. Present R with MODEL label and the no-hearing/no-touch route.
2. **Read · 4 min.** Read all of R; child finds CARD outline, arrow and TAP? question bubble through picture, narration or broad fixed tactile marks.
3. **Model · 5 min.** Teacher says, “The arrow tells me where a suggested action points; the page cannot tell me what anyone heard.”
4. **Child try · 7 min.** Card 71 routes ask for one drawing fact, one unknown, and a question an optional approved gentle action could address.
5. **Check · 4 min.** Offer “The card made a loud sound because the picture says TAP?” Child corrects MODEL / NO SOUND SUPPLIED.
6. **Exit · 2 min.** Save question and source label, not an invented response.

## Week 15 · Day 72 — Predict for a verified broad material

**Goal:** make a bounded before-try prediction for one named verified material/setup without importing a paper result. **Proposed codes:** AC9SFU03, AC9SFI01, AC9SFI02. **Prepare:** [ask-predict](ask-predict.svg), verified approved broad card if available, Model R fallback, Card 72.

1. **Preflight · 3 min.** Adult checks material identity or UNKNOWN, broad clean edge, stable support, allergy/sensory approval and consent. If not fully suitable, choose quiet MODEL ONLY.
2. **Source · 4 min.** Name actual object/support/local date when available; R's CARD label is only fictional and cannot verify material.
3. **Model · 5 min.** Teacher fills question, one very gentle adult action and prediction **before** action; “not discernible” is a reasonable possibility, not failure.
4. **Child try · 7 min.** Card 72 routes speak/point/sign/AAC a possible response and reason. Record exact prior wording and access; no tap is needed to form the question.
5. **Check · 4 min.** Ask what must be recorded later: actual action, observer, date, material, support and response. Do not score confident loudness guesses.
6. **Exit · 2 min.** Leave actual result NOT TESTED until a separately approved optional action.

## Week 15 · Day 73 — One gentle actual action, if welcome

**Goal:** record an exact response from one safe optional adult action, or transparently remain model-only. **Proposed codes:** AC9SFU03, AC9SFI02, AC9SFI03. **Prepare:** [one-action](one-action.svg), approved setup and comfort check, Day 72 prediction or NOT RECORDED, Card 73.

1. **Consent/source · 3 min.** Confirm local approval, verified actual object/support/date, warning and each participant's quiet choice. If anyone needs a silent room, do not tap there; use R.
2. **Prior words · 4 min.** Read true before prediction or NOT RECORDED. An after-event guess cannot become an earlier prediction.
3. **Optional action · 5 min.** Adult makes at most one very gentle fingertip tap at the stated point after warning, then stops. A child may observe directly if comfortable or receive neutral attributed ADULT REPORT; no compulsory touch/hearing.
4. **Exact record · 7 min.** Card 73 routes record local date, object/material, support/action, DIRECT or ADULT REPORT, and exact heard/felt/not discernible response. If no action, write NOT TESTED; describe R only as paper.
5. **Check · 4 min.** Ask whether one response proves every card/fabric behaves this way. Child names setup and one unknown (thickness/support/action).
6. **Exit · 2 min.** Save source-bounded observation privately; do not call it a rhythm or capacity test.

## Week 15 · Day 74 — Which conditions matter to a comparison?

**Goal:** identify conditions that would need to stay similar for a careful material-response comparison, without claiming material alone caused a response. **Proposed codes:** AC9SFU03, AC9SFH01, AC9SFI03. **Prepare:** actual Day 73 record if available, Model S, [one-action](one-action.svg), Card 74.

1. **Source · 3 min.** Retrieve genuine dated Day 73 record or mark NOT AVAILABLE. Model S is FICTIONAL, not prior class data.
2. **Read · 4 min.** Read S's “quiet” story claim and list what is missing: material, support, date, action strength, observer.
3. **Model · 5 min.** Teacher contrasts “one soft response in this setup” with “this material is always quiet”; thickness, support and action can differ.
4. **Child try · 7 min.** Card 74 routes choose two conditions to keep similar in a future approved comparison and one observation that would still remain unknown. No second tap is required.
5. **Check · 4 min.** Offer “S proves fabric is quieter than card.” Child notices S names neither verified material nor matched pair.
6. **Exit · 2 min.** Save a cautious comparison plan or model-only question.

## Week 15 · Day 75 — Fresh action-source check

**Goal:** independently interpret a new paper source about a claimed material response and specify what actual check would be needed. **Proposed codes:** AC9SFH01, AC9SFI01, AC9SFI03, AC9SFI05. **Prepare:** [teacher-held Day 75 source](ASSESSMENT.md), blank [ask-predict](ask-predict.svg), Card 75 **after** check.

1. **Preflight · 3 min.** Teacher checks the unseen card and literal read-aloud/tactile route. No live tap is necessary for this check.
2. **Prompt · 4 min.** Give the complete fresh source once, explicitly paper-only.
3. **Think · 5 min.** Allow independent pause or neutral logged reread; no clue about the claim's limits.
4. **First independent response · 7 min.** Child identifies what is drawn/written, what real sound/material fact is unknown, and one safe question; capture exact route/prompt before feedback.
5. **Bound · 4 min.** Ask what verified object, support, date, comfort check and gentle actual action would be needed. Paper route remains REAL OBSERVATION NOT AVAILABLE.
6. **Close · 2 min.** Use [separate key](TEACHER-KEY.md) for next move; Card 75 practice can follow later.

## Week 16 · Day 76 — What is a layered object?

**Goal:** recognise that one broad card with a broad layer has at least two verified parts if actually assembled, while a paper label alone cannot verify material. **Proposed codes:** AC9SFU03, AC9SFI01, AC9SFI05. **Prepare:** Model L, [layer-compare](layer-compare.svg), optional approved large clean card/layer, Card 76.

1. **Source · 3 min.** Label L FICTIONAL MODEL. Adult may separately show an intact verified broad card and safe broad layer with no tap; if unavailable, use paper only.
2. **Read · 4 min.** Read L fully: A CARD outline; B same outline with a striped LAYER; no actual response is supplied.
3. **Model · 5 min.** Teacher says “This drawing labels two parts. Only a real verified object lets us describe its actual material composition.”
4. **Child try · 7 min.** Card 76 routes identify A/B parts, source and unknown, or record verified real card + layer if actually examined safely.
5. **Check · 4 min.** Ask if the striped drawing proves fabric, softness or quietness. Child distinguishes labels/appearance from verified observed properties.
6. **Exit · 2 min.** State one question for an optional safe single-action comparison, not a sound result.

## Week 16 · Day 77 — Hold the card and support; add one layer

**Goal:** compare composition condition A card alone and B same card plus one broad approved layer, with actual response only if safely observed. **Proposed codes:** AC9SFU03, AC9SFI01, AC9SFI02, AC9SFI03. **Prepare:** [layer-compare](layer-compare.svg), approved verified materials/comfort, Model L, Card 77.

1. **Pair gate · 3 min.** Adult checks same real card, stable support/tap point, broad safe verified layer, warning/consent. If any condition or comfort fails, MODEL ONLY.
2. **Predict · 4 min.** Read a genuine earlier A record if comparable or mark NOT AVAILABLE. Child records before prediction for B; L's arrow supplies no actual response.
3. **Optional actions · 5 min.** If fully approved, adult makes one very gentle tap for A and, after separate comfort check, at most one for B. Stop on distress or change of setup. Never amplify or repeat rapidly.
4. **Child record · 7 min.** Card 77 routes log each actual source/date/material/support/action/direct or ADULT REPORT/exact response; if only one or none occurred mark B or both NOT TESTED.
5. **Check · 4 min.** Ask what changed: added layer also changes thickness/mass/contact, so response difference cannot be attributed to one pure material property.
6. **Exit · 2 min.** Keep cautious A/B statement or MODEL ONLY; capacity remains untested here.

## Week 16 · Day 78 — Can two records make a pattern?

**Goal:** distinguish two comparable actual records from a universal law, and record inconsistent or non-discernible responses honestly. **Proposed codes:** AC9SFU03, AC9SFI03, AC9SFI04. **Prepare:** any approved actual A/B records, Model L, Card 78.

1. **Record gate · 3 min.** Check actual dates, materials, support, action and access. If pairs are incomplete/different, mark INCOMPARABLE; no new sound required.
2. **Read · 4 min.** Adult reads exact A/B responses, including NOT DISCERNIBLE or NOT TESTED. Paper L remains diagram-only.
3. **Model · 5 min.** Teacher says “These records describe two actions here; they do not prove every layered card is quieter.” Compare prediction with later actual observation only if both exist in order.
4. **Child try · 7 min.** Card 78 routes make a same/different/unclear comparison and name one changed condition and one limit; use attributed adult reports appropriately.
5. **Check · 4 min.** Offer “No sound heard means no effect happened.” Child says observer may not discern a response; do not infer physical absence.
6. **Exit · 2 min.** Save one bounded claim or reason evidence is insufficient.

## Week 16 · Day 79 — Explain a composite material claim

**Goal:** share an exact source-bounded composition/response idea and next safe question, without claiming a design or music result. **Proposed codes:** AC9SFH01, AC9SFI03, AC9SFI05. **Prepare:** [explain-next](explain-next.svg), actual private record if available or Model L, Card 79.

1. **Source/privacy · 3 min.** Choose a genuine dated record or L MODEL; teacher may be the only listener. No child audio/photo recording is needed.
2. **Read · 4 min.** Name actual verified parts and exact response/access, or just L's drawn labels/unknown.
3. **Rehearse · 5 min.** Teacher models “In this setup/source I observed/was told … ; I do not know … .” No universal loudness or capacity inference.
4. **Child share · 7 min.** Card 79 routes speak, sign, AAC, draw/point or direct an adult to share the same bounded idea; adult records exact response route.
5. **Feedback · 4 min.** Record a listener's actual words only if they occur; otherwise NO ACTUAL LISTENER RESPONSE. Any revision can come from source review.
6. **Exit · 2 min.** Save one optional future safe material question, not a claim of growth.

## Week 16 · Day 80 — Fresh composite-source check

**Goal:** independently interpret an unseen layered-material source, distinguish representation from actual property and plan a safe check. **Proposed codes:** AC9SFU03, AC9SFI01, AC9SFI03, AC9SFI04, AC9SFI05. **Prepare:** [teacher-held Day 80 source](ASSESSMENT.md), blank [layer-compare](layer-compare.svg), Card 80 only after response.

1. **Preflight · 3 min.** Teacher checks fresh card and access route; no sound action is part of the public paper check.
2. **Prompt · 4 min.** Present the complete source once, explicitly FICTIONAL MODEL.
3. **Think · 5 min.** Give independent time; any neutral reread is logged, not used to teach the answer.
4. **First independent response · 7 min.** Child states drawn parts, what claim the page cannot establish, one controlled safe question and source/access; save exact first wording.
5. **Bound · 4 min.** Ask what actual approved material/action/date and before prediction would be required; without them mark REAL OBSERVATION NOT AVAILABLE.
6. **Close · 2 min.** Consult [separate key](TEACHER-KEY.md). One check cannot certify the term, year or material understanding.