Move through all three source passages without mixing their questions into one contextless pool. Practice structure-first, targeted-search, or hybrid reading under a visible open clock.
The source manifest fixes 16 + 18 + 16 questions. This route may shuffle passage-block order, question order inside a block, and answer positions, but it may not drop a question, duplicate one, or place a question beside the wrong passage.
01
Lake layers
When a Lake Becomes Layers
1,243 words · 16 questions · 8 stable paragraph anchors
16
02
Magnetic map
Two Questions in One Field
1,334 words · 18 questions · 9 stable paragraph anchors
18
03
Clock in a tube
A Clock in a Tube
1,312 words · 16 questions · 9 stable paragraph anchors
16
THREE OPTIONAL WORKFLOWS
Change the route. Not the evidence standard.
Each circuit contains the same fifty frozen questions and three complete passages. The chosen strategy changes the visible scaffold, not the answer key or the claim boundary.
01
Map every passage before its questions
Structure first
Read each passage block once, record the paragraph jobs, then answer its intact question set.
Prepare
Build a light paragraph map before opening the first question in each block.
Navigate
Return through paragraph jobs and transitions rather than searching isolated words.
Prove
Replace every remembered summary with the smallest controlling quotation before checking.
Audit
Mapping time is part of the visible open clock; the map must shorten later returns instead of becoming a second passage.
02
Let the stem choose the first return
Targeted search
Start with each question, extract its locator language, and search the active passage for the smallest complete evidence window.
Prepare
Scan paragraph anchors and distinctive terms without trying to memorize the entire passage.
Navigate
Use names, relationships, experiments, and qualifiers from the stem to choose a paragraph.
Prove
Read around the match until reference, contrast, degree, and condition are stable.
Audit
A keyword hit is a location hypothesis, never the answer by itself.
03
Carry a light map into targeted returns
Hybrid map + search
Orient to the passage structure, then use question-specific locators and switch methods when the evidence spans multiple paragraphs.
Prepare
Name the subject, overall movement, and two or three high-value paragraph landmarks.
Navigate
Search locally for explicit details; reopen the structure map for integration, inference, and author-level questions.
Prove
Join only compatible spans, then test the option against the author’s stated boundary.
Audit
Switching is deliberate only when it improves evidence control; it is not permission to guess from topic familiarity.
OPEN-CLOCK PACE LEDGER
A budget. Never a buzzer.
Sixty minutes divided by fifty questions gives 72 seconds per item. Multiplying that planning average by each frozen passage allocation produces the local budgets below. They are transparent arithmetic—not official per-item or per-passage limits.
The circuit keeps elapsed and visible-item decision time on screen, but it never submits, hides, or locks an answer. Reading and immediate-feedback review make this learning route deliberately different from the completion form.
Nonbinding planning checkpoints for the frozen source order
Passage set
Items
Local budget
Source-order cumulative
Lake layers
16
19:12
19:12
Magnetic map
18
21:36
40:48
Clock in a tube
16
19:12
60:00
Section planning total
50
60:00
Active circuits may move these passage blocks into a different order. Each block keeps its own item count and local budget.
THREE PASSAGES · FIFTY QUESTIONS · ONE OPEN CLOCK
Build the complete retrieval circuit.
Select a strategy before the shuffle. The active passage, its paragraph anchors, and the chosen scaffold stay beside every question; complete evidence appears immediately after checking.
Every circuit contains all fifty source questions. The three passage blocks shuffle as units, questions shuffle only inside their own block, and all five answer choices receive a fresh order.
LEARNING-MODE BOUNDARY
Mixed retrieval is not a completion form.
This route preserves deterministic three-block assembly, all three optional strategy paths, six-objective result grouping, visible open-clock decision timing, local recovery, and exact evidence feedback without changing any passage, question, option, or answer identity.
It does not enforce the official 60-minute limit, delay feedback, create raw form reporting, claim score conversion, establish device evidence, run a passage-aware pilot, or approve the Reading release.
The completion form provides the official 50-question/60-minute and untimed modes with exact recovery, delayed review, and raw reporting. This learner route is open as a public beta while real rights, scientific, passage, assessment, accessibility, and calibration evidence remains open.