# Foundation / Prep Science · Term 2 Weeks 19–20

**Teacher draft · 10 distinct 25-minute lessons, Days 91–100.** Week 19 asks what a careful movement comparison can and cannot tell us. Week 20 turns a dated, source-labelled Science record into a private explanation and next question. This is a Science companion to the integrated [Week 19](../../../term-2/week-19-what-a-test-tells-us.md) and [Week 20](../../../term-2/week-20-teach-someone-else.md) guides. The physical example here uses one large closed carton on its broad or narrow face on a shallow adult-controlled tray; it does **not** reuse the integrated ring and paper-surface trial. Do not run both physical tests merely to fill a timetable. Choose the approved local setup and teach the same reasoning with the paper-only route if no physical investigation is safe or available.

Start with [materials and complete source texts](MATERIALS.md), then [daily scripts](LESSONS.md), [learner cards](LEARNER-CARDS.md), [optional worked swaps](OPTIONAL-PRACTICE.md), [family bridges](FAMILY-OPTIONAL.md), and [print/text/tactile aids](#original-a4-aids). Keep [fresh checks](ASSESSMENT.md) unseen until the first independent response; use the [separate key](TEACHER-KEY.md) afterwards. [National/Queensland crosswalk](CURRICULUM-CROSSWALK.md), [source and rights ledger](SOURCE-RIGHTS.md), and [honest desk review](QA-RUN-THROUGH.md) state the limits.

## Non-negotiable evidence boundary

- **REAL OBSERVATION NOT AVAILABLE** is the result when the school has no approved, actually run, dated setup. A page model cannot show that a carton really slid, a material had a property, or a child made progress. Record actual equipment, source/date, access, prediction timestamp/order, each trial, interruptions and exact adult support. Do not retrofit a prediction after a result.
- A person can watch, listen to a neutral attributed description, use a broad secured tactile diagram, AAC/sign/pointing, or direct an adult scribe. These routes pursue the same question. **ADULT REPORT** is attributed information, not the child's direct sensory observation. No fixed learning-style label.
- Closed large carton only if school-approved: no staples, sharp edges, unknown fillings, small parts, elevated launches, fast objects or animal interaction. Adult controls the pretested shallow rimmed tray; children need not touch, hear or watch the test. A quiet paper route is always available and marked **FICTIONAL MODEL**.
- No child names, photos, family data, precise locations or private support details enter the public assets. Optional home prompts are conversation starters, not homework or a required home experiment.

## Original A4 aids

Each linked vector has a one-page selectable-text PDF and a numbered exact-text/tactile construction route in [MATERIALS.md](MATERIALS.md). Blank fields are intentionally blank; none claims an actual run.

| Aid | Print |
| --- | --- |
| Question and prediction mat | [SVG](question-predict.svg) · [PDF](question-predict.pdf) |
| Safe setup and fair-change map | [SVG](setup-change.svg) · [PDF](setup-change.pdf) |
| Four-run raw record | [SVG](four-run.svg) · [PDF](four-run.pdf) |
| Interruption and claim sorter | [SVG](claim-limit.svg) · [PDF](claim-limit.pdf) |
| Source and evidence portfolio | [SVG](source-portfolio.svg) · [PDF](source-portfolio.pdf) |
| Private handover and next question | [SVG](handover.svg) · [PDF](handover.pdf) |

**Status:** original SubjectNest draft; adult educator, accessibility, local syllabus, equipment/safety and actual classroom review pending. The deterministic [verifier](verify_pack.py) checks structure, source rows, links, asset text and file hashes, not pedagogical effectiveness. No classroom pilot or child result is claimed.

Original scripts, fictional data, cards and vectors © 2026 NeuroForgeIO Pty Ltd, [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Attribute SubjectNest by NeuroForgeIO Pty Ltd, link source/licence and note changes. ACARA retains its own rights in curriculum wording.