Unseen Text Reading Strategy | Build the Passage Model Before You Hunt Answers

21. A second read should have a reason

The second read is not automatically “read everything again more carefully”. It should repair or deepen the map. Perhaps the first pass established movement but not viewpoint. Perhaps one paragraph remained unstable. Perhaps the task requires close attention to a quoted phrase. Give the reread a purpose before beginning it.

Purposeful rereading is faster because the reader knows what counts as success. “I need to locate where the writer becomes less certain” has an exit condition. “I should read it again” does not. The second approach often becomes an anxiety ritual: the learner reaches the bottom with no new decision and starts again.

Leonie times the difference during practice. Targeted rereads are short and produce a change in the model. Broad rereads often feel safer while quietly consuming the minutes she needs later.

22. Decide whether questions should be previewed before the first read

Some examination formats reward a quick look at the questions because the tasks establish useful search targets. Others contain enough questions, subparts or wording that previewing fragments attention before the passage has shape. There is no universal rule that question-first or passage-first is always superior.

Test the decision under the actual paper design. If previewing questions causes Iona to hunt only for isolated answers and miss the passage movement, the preview is too expensive. If reading the whole passage without knowing any task causes repeated returns to details that could have been anticipated, a light preview may help.

The principle is stable: whichever sequence you use, do not let the task destroy the text model. A question-led read still needs orientation; a passage-led read still needs later task precision.

23. Annotation is an external memory system, not proof of effort

Annotation is useful when it stores something the reader would otherwise have to rediscover. A bracket around a turn, a word identifying paragraph function, an arrow showing cause, or a note marking a changed viewpoint can reduce later search. Underlining whole sentences because they “look important” often cannot.

Every annotation has a cost: time, visual clutter and divided attention. The correct question is not “Should students annotate?” but “Which mark changes a later decision enough to repay its cost?” Different readers may keep different systems. Iona may benefit from evidence brackets; Leonie may need sparse margins because too many marks make the page feel busier and increase her time anxiety.

Keep only annotations that improve retrieval, interpretation or control.

24. Do not annotate devices before you know what the text is doing

A common unseen-text routine is to circle metaphors, underline adjectives and label techniques immediately. This can produce a page full of features before the reader knows the controlling movement. The features then float without purpose.

Read for meaning first. Once you know that the scene is moving from ordinary routine toward threat, repeated sound imagery may matter because it drives that movement. Once you know that the writer is complicating an optimistic claim, a shift from confident verbs to hedging becomes meaningful. Device recognition should serve interpretation rather than replace it.

Maren asks a simple question before labelling a technique: “What change in meaning, pace, stance or reader attention does this help create?” If she has no answer, the label can wait.

25. Build a map that predicts evidence locations

A useful passage model is predictive. If the question later asks for the first reason the writer becomes doubtful, the reader should have a rough idea where doubt entered. If the question asks how a character’s attitude changed, the reader should know which sections represent before, pressure and after.

This is why paragraph-job labels outperform topic labels. “Example of public resistance” tells the reader where resistance evidence lives. “Paragraph about transport” does not. The map reduces the search space from the entire text to a likely corridor.

The examination advantage appears later, but it is earned during the first read. A good map turns future evidence search into navigation rather than rediscovery.

26. Preserve unresolved questions without letting them block progress

Strong readers sometimes misunderstand their own uncertainty. They assume every unresolved detail must be solved before moving on. In an unseen text, some uncertainty is normal and may be resolved by later information.

Mark uncertainty lightly: “who is ‘they’?” “why does she refuse?” “writer serious or ironic?” Then continue if the missing detail does not prevent understanding of the next relationship. Later sentences may answer the question. A task may make the uncertainty relevant. Or it may never matter.

Iona learns the difference between productive uncertainty and structural collapse. If she still knows what the paragraph is doing, she can continue. If she no longer knows who is acting or what claim is being discussed, repair immediately.

27. Use contradiction as a signal to update the model

When a later sentence seems to contradict the model, do not automatically discard the sentence or force it into the earlier interpretation. Ask whether the writer has changed direction, introduced another viewpoint, qualified the claim or revealed that the earlier assumption was too broad.

This is particularly important in literary passages where narrators may misread events, and in arguments where a concession can sound like reversal. The contradiction may be the most informative moment in the passage because it reveals complexity.

Maren writes a tiny question mark beside genuine model conflict. Her job is not to erase the conflict immediately but to identify what changed. That habit prevents early interpretations from becoming prisons.

28. Distinguish the writer’s claim from reported claims

Articles often contain several voices: experts, critics, officials, witnesses, companies, campaigners, researchers. The writer may quote a claim in order to support it, challenge it, frame it neutrally or expose its limitation. A first read that blends all voices into one will make later questions unstable.

Use attribution mentally: the minister claims, the researcher finds, the writer questions, the resident reports. Then ask what the main writer does with each voice. Does the quotation carry authority? Provide contrast? Supply a human example? Become the object of criticism?

This separation is essential in multi-source and media-literacy tasks, but it also matters in ordinary comprehension whenever quotation and commentary interact.

29. In narrative, separate what happens from how it is perceived

A first-person narrator can report direct experience while misinterpreting another person’s motive. A limited third-person narration can colour events through one character’s fear or hope. Unseen reading becomes stronger when the learner distinguishes event from perception.

Suppose the narrator says, “He ignored me because he was furious.” The observable event may be that he did not respond. The fury is the narrator’s explanation unless other evidence confirms it. Later questions about reliability, inference or viewpoint may depend on preserving that distinction.

Iona’s map can include two layers: what occurred and how the focal character interprets it. This keeps literary nuance without requiring premature judgement about whether the narrator is reliable overall.

30. In informational texts, separate mechanism from example

Explanatory writing often introduces a mechanism, then illustrates it. Readers can remember the vivid example and lose the general process. Later questions may ask for the mechanism rather than the case.

Attach the example to the rule it serves. If a paragraph describes one city’s flood barrier after explaining how wetlands slow water, the city may demonstrate the mechanism rather than define it. A sentence-level note such as “example of buffering” keeps hierarchy intact.

This also protects summary work. Examples are often expendable when the task asks for main ideas; mechanisms are not.

31. In persuasive texts, track audience pressure

Persuasive texts do not merely contain arguments; they position readers. Direct address, shared values, fear, reassurance, moral language, concessions and calls to action can change how the audience is invited to respond.

During the first read, do not annotate every persuasive technique. Ask how the relationship with the audience changes. Does the writer begin by establishing common ground, then increase urgency? Does an apparently balanced concession make the final recommendation more credible? Does a personal story reduce distance before a policy claim?

The map should capture rhetorical movement, not a catalogue of devices.

32. Tone is often a trajectory, not a label

Students often hunt for one tone word to describe an entire passage. Many unseen texts move: amused to critical, confident to cautious, detached to personal, hopeful to resigned. The movement can matter more than the starting label.

Track evidence of tone through diction, qualification, sentence shape, punctuation and selection of detail. Then ask whether later sections preserve or alter the stance. A trajectory such as “initially playful, increasingly uneasy” gives later questions a stronger map than “tone = worried”.

Do not over-label minor fluctuations. Look for changes that affect interpretation or structure.

33. Pace is part of the model

Narrative and descriptive texts can accelerate or slow through sentence length, action density, detail, dialogue, interruption and reflection. Pace helps the reader see where attention is being controlled.

A passage may spend half a page on one quiet observation, then compress several events into two sentences. That change is structural evidence. The first read should notice where time expands or contracts, especially if the shift coincides with emotional or narrative importance.

Maren marks pace only when it changes what the passage is doing. The goal is not to count sentence lengths but to understand the allocation of narrative attention.

34. Repetition can be a navigation signal

Repeated words, images, ideas or sentence patterns can connect distant parts of a text. The first occurrence establishes something; a later recurrence may reinforce, transform or reverse it. Recognising the connection can reveal development.

For example, a repeated reference to a closed window may begin as ordinary setting and later become associated with confinement. An argumentative phrase repeated in the conclusion may return with narrower certainty. The repetition creates a bridge across the passage.

Record repeated elements only when they appear to structure meaning. Not every repeated noun is symbolic or deliberate. Evidence of transformation is stronger than repetition alone.

35. A useful map distinguishes core, support and ornament

Not every sentence deserves equal memory. Core propositions carry the main movement. Supporting material explains, evidences or exemplifies them. Ornament may create atmosphere, texture or voice without being central to every question.

This hierarchy is provisional because a later question may make an apparently decorative phrase important. But on the first read, the learner benefits from knowing what must be retained even if detail fades. “The writer’s main claim changes after the study is introduced” matters more globally than the exact name of the study unless a task later targets it.

A passage map is selective by design. Selection is what makes it usable under time.

36. Read images and layout as part of the passage model

Unseen comprehension may include photographs, diagrams, captions, typography, headlines, charts or visual spacing. These are not decorations outside the text. They can establish audience, emphasis, sequence, contrast or evidence.

Start by asking what proposition the visual contributes. Does a photograph personalise a statistical argument? Does a chart support the writer’s scale claim but not the causal explanation? Does a caption narrow how an image should be interpreted? Does layout separate viewpoints that prose alone might blur?

The same modelling principle applies across modes: identify function and relationship. Do not simply list visual features. A colour, image or graph matters because of what it helps the text do.

37. A graph changes the evidence map

Graphs and tables carry their own reading demands. Check variables, axes, units, scale, time period and comparison groups before accepting a headline interpretation. A rising line can represent a small absolute change. A percentage increase can coexist with a falling total. A chart can show association without establishing cause.

When prose and data appear together, map the relationship. Is the writer accurately describing the data? Is the chart an example, the main evidence or a limitation? Does the prose make a stronger claim than the visual supports?

Iona treats the visual as another source with its own proposition. That prevents the common mistake of absorbing the writer’s description of a graph without checking the graph itself.

38. Multiple texts require separate maps before synthesis

When two or more unseen sources appear, resist immediate blending. Build a compact identity for each: source, position, evidence, tone, purpose, important limitations. Only then compare.

Two texts can use similar vocabulary while making different claims. They can appear to disagree while answering different questions. One may describe what happened; another may argue why it happened. Without source separation, the learner can attribute evidence to the wrong writer or manufacture a conflict that is not present.

A good paired-text map might read: “A: optimistic about potential, evidence = pilot study, acknowledges cost. B: sceptical of scaling, evidence = implementation failures, accepts local benefit.” Synthesis can then begin from actual positions.

39. Comparison should be postponed until each side is stable

Early comparison can distort the individual texts. Once the learner notices that Source A is hopeful, she may search Source B only for pessimism. That creates a false binary and ignores nuance.

First ask what each source does on its own terms. Then choose a shared axis: degree, cause, evidence, audience, solution, attitude, method or outcome. Compare positions after each map is sufficiently stable.

Maren finds this especially useful in paired argument. Separate reading protects fidelity; later comparison creates structure. The order matters.

40. Background knowledge should support, not colonise, the text

Knowledge helps readers interpret unfamiliar material. It can clarify terminology, activate likely relationships and reduce cognitive load. But strong prior beliefs can also cause the reader to substitute what she knows for what the passage says.

During an unseen examination, treat the text as the primary model unless the task explicitly invites wider knowledge. If a passage about climate policy uses a narrower definition than the one the learner knows, answer from the source. If a historical account makes a limited claim, do not “correct” it by importing unrelated facts unless evaluation is required.

Iona uses background knowledge as a hypothesis generator, not as evidence. The passage still decides what can be claimed.

41. Familiar topics can create more errors than unfamiliar ones

An unfamiliar topic makes the reader cautious. A familiar one can make her fast and careless. She predicts what the writer “must” mean, skips qualifications and answers from memory.

Training should therefore include topics learners know well. Ask them to distinguish the passage’s exact proposition from their own knowledge. When the text disagrees with expectation, require the model to follow the text rather than the student’s preferred account.

This is a useful transfer test. Genuine reading control works when knowledge is sparse and when knowledge is abundant.

42. Do not guess the author, era or hidden context unless the paper gives a reason

Unseen texts can tempt students into speculative context. A style feels old-fashioned, so they invent a period. A concern sounds political, so they guess an ideology. These guesses can distort the reading and waste time.

Use contextual information supplied by the examination: title, date, source label, introductory note, caption or author description. Beyond that, prefer textual evidence. If the language suggests formality, age or social distance, describe the effect without pretending to know a history the paper has not established.

The discipline is simple: interpret what is available. Unseen reading rewards portable analysis, not confident invention.

43. Titles are hypotheses, not summaries

A title can orient the reader, create expectation, frame a theme or become ironic by the end. Do not assume it tells you exactly what the passage will say. Read it as an initial hypothesis.

After the first pass, return mentally to the title. Has its meaning changed? Does it name the topic, the argument, a metaphor, a quotation, a question or a contrast? A title that seemed literal may become ironic. A broad title may narrow into one surprising case.

That change can become part of the passage map because it reveals how the text controls expectation.

44. Openings tell you what kind of attention the text wants

An opening can begin with action, description, a statistic, a question, a claim, a quotation, an anecdote or direct address. Each creates a different entry condition for the reader.

Ask what the opening establishes and what it withholds. A narrative may begin in the middle of action and delay explanation, forcing the reader to share confusion. An argument may begin with a personal case before revealing the broader issue, making an abstract problem human. A question may invite the reader into shared uncertainty.

Do not overanalyse the first sentence before reading on. The opening’s function often becomes clear only after the later movement is known.

45. Conclusions reveal what the writer wants retained

The ending may resolve, qualify, reopen, intensify or redirect the passage. In argument, it often reveals the final strength of the claim after evidence and concessions. In narrative, it can reframe what came before. In informational writing, it may shift from mechanism to implication.

Compare the ending with the opening. What survived? What changed? Did the writer become more certain or less? Did a repeated image acquire new meaning? Did a character move, or merely understand differently?

A first-read map that includes opening and ending creates a strong axis for questions about development and purpose.

46. Negative space can matter

Texts create meaning through omission as well as statement. A narrator avoids naming an emotion. An argument presents benefits without costs. A character answers every question except one. A source reports percentages without sample size.

Do not turn every absence into significance. Ask whether the omission is made visible by expectation, contrast or consequence. If the missing information changes how confidently a claim can be interpreted, it belongs in the model.

Advanced unseen reading often depends on noticing what the text encourages the reader to expect and then withholds.

47. Irony begins with mismatch

When literal wording does not fit the situation, tone or surrounding evidence, consider whether the text is creating irony. The task is not to label sarcasm whenever a sentence sounds exaggerated. It is to identify the mismatch and explain why the literal reading is unstable.

Signals can include context, repetition, quotation marks, exaggerated praise after obvious failure, or a narrator whose words conflict with events. No single signal proves irony. Converging evidence does.

Iona records the mismatch first: “words praise; events undermine”. The interpretive label can follow once the pattern is secure.

48. Figurative language should be translated back into the model

A metaphor matters because it imports qualities from one domain into another. If the writer describes bureaucracy as a maze, the image may suggest complexity, disorientation and difficulty escaping. Which qualities matter depends on context.

During the first read, do not stop to write a full analysis. Translate the figure into a brief model update: “system framed as confusing and trapping”. This preserves meaning without derailing pace.

Later, if a question targets language effect, return to the exact wording and develop the analysis. First-pass reading should capture function at the level necessary for navigation.

49. Description can function as evidence

Descriptive details are not always ornamental. The condition of a room, the pattern of weather, a repeated physical gesture or the selection of sensory information can reveal character, mood, social condition or change.

Ask why this detail is here relative to the passage movement. A cracked cup may be irrelevant texture in one story and a recurring sign of neglect in another. The distinction comes from pattern and consequence, not from the object itself.

Maren avoids symbolic over-reading by waiting for reinforcement. One detail invites a possibility; repetition or structural placement strengthens it.

50. Dialogue creates a second layer of structure

In narrative unseen texts, dialogue can reveal power, avoidance, misunderstanding, social distance and change. Track not only what is said but who controls the exchange, who interrupts, who answers indirectly and where silence appears.

A conversation may move from routine to confrontation, from unequal to balanced, or from open to evasive. That movement can form the structural spine of an extract even when the setting barely changes.

For the passage map, compress dialogue into relationship changes rather than trying to remember every line. “Mother questions → daughter deflects → silence → daughter finally admits” is a navigational model that later evidence search can use.

51. Silence and interruption can change the relationship map

What characters do not say can matter when the text makes the omission visible. A conversation that has moved rapidly may stop after one question. A character who usually answers directly may evade. A repeated interruption may establish dominance. These patterns belong to the passage model because they alter the relationship between speakers.

Do not infer hidden motives from silence alone. First identify the observable pattern. “He does not answer this question although he answered the previous ones.” Then inspect context before moving to a conclusion such as discomfort, concealment or uncertainty.

This evidence-first order keeps literary interpretation disciplined and gives later inference questions a reliable foundation.

52. Time shifts should be mapped explicitly

Narrative extracts may move between present action, memory, anticipation and reflection. Informational texts may shift from historical background to current evidence to future implication. If the reader loses the time frame, cause and sequence become unstable.

Mark major temporal transitions mentally: now → memory → now; past pattern → current change → projected consequence. Ask what the shift contributes. A memory may explain a present reaction. Historical context may show that a supposedly new problem has older roots. A future projection may transform description into warning.

The timeline does not need every date. It needs enough structure to keep events and claims in the correct order.

53. Spatial movement can be structural too

A narrative may move from public to private space, indoors to outdoors, crowd to isolation, familiar place to unfamiliar territory. These physical changes can accompany changes in emotion, power or knowledge.

Do not assume space is symbolic merely because it changes. Ask whether the move alters what the character can see, whom they can speak to, what they are allowed to do or how the scene is framed. If the consequences are meaningful, the spatial shift belongs in the passage model.

A short note such as “public station → empty platform → enclosed carriage” may become useful if later questions ask how isolation is built.

54. Lists reveal category and emphasis

Lists can accumulate evidence, accelerate pace, create excess, organise categories or overwhelm the reader. The first-read question is not “there is a list” but “what role does the accumulation play?”

In an argument, a sequence of costs may make scale visible. In description, sensory details may create richness or claustrophobia. In a speech, repeated examples may build inclusiveness by moving through different groups. The surrounding movement determines the function.

Maren compresses a long list into its category when mapping: “examples of hidden costs” is usually more useful than memorising six items on the first pass.

55. Contrast is one of the highest-value structural signals

Many comprehension questions are built around contrast: expectation versus reality, past versus present, public appearance versus private thought, claim versus evidence, one source versus another. A first read that notices contrast already predicts several possible question jobs.

Record both sides. “Officials call project efficient / residents describe delay.” “Character says she is calm / behaviour becomes increasingly restless.” “Opening celebrates convenience / ending questions dependence.” Contrast is not just the word “but”; it can be distributed across sections.

Because contrast creates relational memory, it is easier to retrieve later than isolated facts.

56. Concessions are not surrender

Argumentative writing often concedes a point without abandoning the main position. “The scheme is expensive, but…” does not mean the writer is against it. The concession may increase credibility by acknowledging a weakness before defending a stronger claim.

Map concession as relationship: accepts X, still concludes Y. Later questions about attitude, balance or evaluation depend on preserving both parts. Learners who remember only the concession can invert the writer’s stance.

Iona marks the pivot rather than the negative detail alone. The main question is what survives after the concession.

57. Definitions establish local rules for the passage

Informational texts may define a term in a specialised way. That local definition controls later meaning even if the learner knows a broader everyday use. Missing the definition can destabilise the whole model.

Notice phrases such as “by this we mean”, “defined here as”, “refers to”, appositive explanations and examples that establish category boundaries. Keep the definition available when the term returns later.

This is one reason unfamiliar vocabulary should not always be inferred from general knowledge. The writer may be constructing a precise local meaning that matters more than the learner’s prior definition.

58. Questions inside the passage can organise the argument

Rhetorical or genuine questions can mark a transition. A writer may ask, “So why has adoption remained slow?” and then spend the next section answering it. That question becomes a structural signpost.

Map the question and answer relationship. Does the question introduce the central problem, challenge an assumption, involve the reader or signal a shift from description to explanation? The function depends on what follows.

Do not automatically call every question rhetorical. If the text genuinely explores the answer, treating it as a structural problem statement is more accurate.

59. Examples of failure often carry more structure than examples of success

Writers use failure cases to expose hidden conditions. An article may praise a technology, then describe where it failed, revealing that the earlier claim works only under certain circumstances. A narrative may show a plan going wrong, forcing a character to change.

When a failure example appears, ask whether it merely illustrates difficulty or actually changes the model. Does it introduce a limitation? Reveal a character trait? Create the turn? Establish a new causal factor?

These moments are disproportionately useful because they often transform general understanding into conditional understanding.

60. Conditional language tells you when a claim applies

If, unless, provided that, only when, except where and similar structures define the boundary of a claim. Missing the condition can make the passage look more certain or universal than it is.

When the text says a method works “only when trained staff are present”, the staff condition belongs to the model, not to minor detail. Any later summary of effectiveness that omits it changes the proposition.

Advanced readers learn to see conditions as structural constraints. They answer not merely “what is true?” but “under what conditions is the writer saying it is true?”

61. Quantifiers control how big the claim is

Some, many, most, few, often, rarely, typically, entirely and partly are model words. They tell the reader how widely a statement applies. A passage can be misunderstood while every content word is recognised if the quantifier is lost.

During a fast first read, resist the temptation to compress “many respondents” into “respondents” or “occasionally fails” into “fails”. The simplified memory may become false. Keep high-value scope words attached to the claim.

Iona sometimes circles only the quantifier rather than the whole sentence because that one word controls the degree later answers must preserve.

62. Negation deserves disproportionate attention

Not, never, no longer, hardly, rarely, neither and except can reverse or sharply narrow meaning. Under time pressure, readers sometimes retain the topic and lose the negative force.

A sentence such as “The evidence does not show that the programme fails” is not equivalent to “the evidence shows that the programme succeeds”. The negation removes one conclusion without proving its opposite. That distinction can become central in evaluation or inference questions.

When a difficult sentence contains negation, rewrite it mentally in simpler form without changing logical strength.

63. Absence of evidence is not automatically evidence of absence

Texts sometimes report that no effect was detected, no record exists or no difference was found. Readers may turn this into proof that the effect, event or difference does not exist. The conclusion can be stronger than the source permits.

Map exactly what was observed. “Study did not detect a difference” may indicate limited evidence, not certainty of equality. “No record was found” may mean absence from the available archive, not proof the event never happened.

This distinction is especially valuable in informational and argumentative unseen texts because it protects against overclaiming before any question is asked.

64. Causal claims need mechanisms or evidence

A passage may report that two things occur together without stating that one causes the other. Strong first-pass reading preserves that uncertainty. Look for causal verbs, temporal sequence, proposed mechanism, experimental comparison or explicit qualification.

If the writer says increased screen use “was associated with” lower sleep duration, do not silently upgrade the model to “screen use caused less sleep”. The passage may later discuss alternative explanations. Keeping association separate from causation makes that later section easier to integrate.

Unseen comprehension rewards logical fidelity before it rewards sophisticated vocabulary.

65. Model the writer’s burden of proof

Different claims require different evidence. A writer claiming that one person had a bad experience needs less evidence than a writer claiming that a policy harms an entire population. As the claim widens, the burden of support rises.

During an unseen argument, compare scale of claim with scale of evidence. Is one anecdote being used to establish a universal conclusion? Is a small pilot study framed cautiously or presented as decisive? Does the writer acknowledge limits?

This evaluative awareness can sit quietly inside the passage model. The reader does not need to judge everything immediately, but should know where support is strong, narrow or contested.

66. Read statistics according to what they can actually prove

Numbers can feel definitive, which makes them easy to overread. A percentage tells you proportion but not necessarily sample size. An average can hide variation. A dramatic increase may begin from a tiny baseline. A survey reports responses, not automatically objective behaviour.

During a first read, do not perform a full statistical critique unless the task demands it. Instead, map the function and the obvious limits: “survey used to show public concern”, “small study supports possibility, not certainty”, “historical trend establishes scale”.

This keeps the evidence useful without allowing numerical appearance to exceed evidential strength.

67. Expert authority is part of the source map, not an automatic guarantee

Writers often quote specialists to establish credibility. The first-read question is what authority the source contributes. Is the expert speaking within a relevant field? Reporting research? Offering interpretation? Representing an institution with interests in the issue?

Do not dismiss authority merely because bias is possible, and do not accept every expert quotation as final. Map role, relevance and function. “Engineer explains mechanism” is different from “company spokesperson defends policy”.

Later evaluation questions can deepen the judgement. The passage model only needs enough source separation to know whose evidence is doing what.

68. Anecdotes humanise; they do not automatically generalise

A single story can make an issue concrete, create empathy, establish stakes or introduce a problem. It can also function as evidence of one experience. What it cannot do alone is establish how common that experience is.

Map the anecdote according to function. “Opening story makes cost personal.” “Case illustrates failure mode.” “Personal memory establishes narrator’s connection.” If the writer later moves to broader data, note the shift from example to general support.

This hierarchy makes unseen arguments easier to remember and later evaluate.

69. Loaded language can reveal framing before it reveals truth

Words such as reckless, visionary, wasteful, courageous, merely or even can position the reader before evidence is assessed. A first read should notice framing without assuming that emotionally strong language makes the underlying claim false.

Ask what viewpoint the wording encourages. Does it make one option appear normal and another extreme? Does it assign blame? Does it minimise a counterargument? This helps map stance and audience positioning.

Iona separates two questions: how is the issue framed, and what evidence supports the claim? Keeping them separate prevents rhetorical force from being mistaken for proof.

70. Headings and subheadings reveal hierarchy

In articles and informational texts, section headings tell the reader how the writer has divided the problem. They can indicate sequence, categories, questions, contrast or progression from diagnosis to solution.

Use headings as navigation, not as substitutes for reading. A heading called “The hidden cost” predicts a negative development, but the paragraph determines what the cost is and how strongly the writer treats it. Headings can also be playful or ironic.

A compact map can borrow the hierarchy: problem → hidden cost → alternative → limitation → conclusion. This gives the reader a retrieval scaffold almost for free.

71. Sentence length matters when it changes processing or pace

Long sentences can accumulate qualification, delay the main clause or imitate crowded thought. Short sentences can isolate a conclusion, create impact, accelerate action or interrupt an established rhythm. But length alone is not meaning.

On the first read, notice a sentence when its shape changes the movement. A sudden fragment after long explanation may signal shock or certainty. A long sentence after clipped dialogue may open reflection. These shifts help locate structural emphasis.

Maren avoids the formula “short sentence = tension”. She asks what changes relative to the local pattern and why that change matters.

72. Punctuation should be read through syntax and voice

Dashes, colons, semicolons, brackets, ellipses and exclamation marks can organise thought, interrupt it, add qualification, create delay or foreground an aside. Their effect depends on grammar and context.

Rather than labelling punctuation as expressive by default, ask what relationship it creates. A colon may announce explanation. Dashes may insert a disruptive qualification. Ellipses in dialogue may suggest hesitation, interruption or trailing thought depending on the surrounding exchange.

For passage modelling, the priority is whether the punctuation signals a change in direction, voice or pace worth remembering.

73. Pronoun choice can map social distance

We, they, you, those people and one can position groups and readers. An argument may begin with inclusive “we” and later separate “they”, constructing shared identity before contrast. A speech may use “you” to create responsibility or intimacy.

Do not treat pronouns as automatically persuasive. Map what groups they create and whether those groupings change. Who is included? Who is excluded? Does “we” refer to everyone, the writer and audience, or a narrower professional group?

This social map can become central in questions about audience, viewpoint and rhetoric.

74. Modality is the grammar of confidence and obligation

Must, should, could, may, will and might encode different levels of necessity, recommendation, possibility and prediction. A passage can shift stance through modality even when topic remains constant.

If a writer moves from “governments could act” to “governments must act”, urgency has increased. If a proposal “will solve” becomes “may help”, confidence has narrowed. These changes belong in the model because later tone and purpose questions may depend on them.

Iona often notices modality before adjectives because it gives a cleaner measure of how strongly the writer commits.

75. Active and passive voice change agency

“The committee rejected the proposal” names an agent. “The proposal was rejected” foregrounds the event and can leave the agent unstated. Passive voice is not automatically evasive; sometimes the actor is unknown, obvious or irrelevant. But agency matters to interpretation.

When a passage repeatedly removes agents from controversial actions, the pattern may matter. When scientific writing uses passive structures to focus on procedure, that may reflect genre rather than concealment.

Map who acts, who receives action and when agency disappears. This helps with questions about responsibility and emphasis without forcing a political reading onto every passive sentence.

76. Lexical fields are useful only when they track a pattern

A cluster of words related to war, machinery, illness or nature can frame an issue consistently. One isolated word rarely justifies a broad lexical-field claim. Look for recurrence and distribution.

If a policy article repeatedly uses battle language—fight, defend, attack, frontline—the pattern may frame disagreement as conflict. If one word appears once, treat it locally until more evidence arrives.

For first-read mapping, summarise the pattern’s function: “conflict language intensifies debate”. Save detailed quotation analysis for a later question if needed.

77. Sensory detail can organise attention

Writers choose what the reader sees, hears, smells, tastes or feels. Repeated attention to one sense can shape atmosphere and viewpoint. A narrator who notices every small sound may be alert, anxious or simply placed in a quiet environment; context decides.

Track sensory selection as pattern rather than isolated technique. Does sound increase as threat approaches? Does visual detail disappear when a character becomes disoriented? Does tactile description make a place feel immediate?

A sensory map can become a structural map when changes in attention coincide with changes in the scene.

78. Refrain from solving symbolism too early

Unseen literary texts encourage symbolic interpretation, but early certainty is risky. A river, door, bird or colour can carry thematic weight, or it can simply belong to the setting. Meaning grows through recurrence, placement, contrast and connection to character or theme.

On the first read, record the concrete pattern: “door repeatedly closed/opened at moments of decision”. Do not immediately declare “door symbolises opportunity”. If the later movement supports that reading, the interpretation becomes stronger.

Evidence should make symbolism emerge, not symbolism make evidence disappear.

79. Identify the passage’s centre of gravity

Not every paragraph carries equal conceptual weight. Some unseen texts have a centre of gravity: the paragraph where the problem is reframed, the moment where a character understands something, the section where competing evidence is resolved.

Ask which part most changes the meaning of what comes before and controls what comes after. This is not always the longest paragraph or dramatic event. It may be a quiet qualification that narrows the entire argument.

Finding the centre of gravity helps later because difficult questions often target exactly the text’s most consequential shift.

80. A map should survive when the text is covered

After reading, cover the passage briefly and reconstruct its movement. Can you state the starting position, main stages, important turn, final position and one or two uncertainties? Can you locate roughly where each part sits on the page?

This is not a memory contest. If exact wording disappears but relational structure remains, the map is working. If the reader remembers several vivid details but cannot say how the text develops, the first read collected fragments rather than architecture.

Practice this retrieval for thirty seconds after passages. It strengthens the habit of reading for organisation rather than accidental memorability.

81. The first-read clock should be calibrated, not worshipped

There is no universal number of minutes that every secondary learner should spend on an unseen text. Passage length, density, paper design, reading speed and question weighting differ. A rigid timing rule can make a careful reader rush or give an inefficient reader permission to over-read.

Calibrate timing from realistic practice. Measure how long a successful first read takes when the map is good enough to support later questions. Then compare with unsuccessful passes. Did more time produce a better map, or merely more annotation? Did less time preserve structure but lose key qualifications?

Leonie’s target is not “read in four minutes”. It is “reach a functional map within a sustainable share of the paper”.

82. Reading faster is often a search problem, not an eye-speed problem

Students who say they read too slowly may actually be rereading too broadly, stopping at low-value vocabulary, annotating too much or restarting after every uncertainty. Their basic reading rate may be adequate while their decision process is expensive.

Audit where the minutes go. Time the uninterrupted first pass separately from later evidence search. Record how often the learner returns to the top. Count annotations only if they are genuinely being used. Ask what each pause was trying to resolve.

A passage map reduces later search, which can make the whole section faster even if the initial read itself does not accelerate.

83. Skimming and close reading are different gears

Skimming is useful for orientation, structure and location; close reading is useful when exact wording, syntax, implication or effect matters. Problems arise when learners use one gear for the entire task.

A first orientation may move relatively quickly through examples while slowing at the main claim or turn. A later language question may demand close attention to three words. Efficient reading changes resolution according to the job.

Iona imagines zoom levels rather than speed labels. Wide view for movement, medium view for paragraph relationships, close view for the evidence corridor. The ability to change scale is more useful than trying to read everything at one uniform intensity.

84. The reader should know why she is slowing down

Slowing down is productive when the sentence carries a turn, definition, qualification, dense syntax or evidence central to the passage. It is unproductive when the learner stops because a word looks sophisticated but does not affect the model.

During practice, ask learners to mark where they slowed and why. The record reveals whether attention is allocated to high-value decisions. Some students repeatedly slow at names, numbers and unfamiliar nouns while moving too quickly through contrast and negation.

Good unseen reading is selective attention. The page does not deserve equal processing everywhere.

85. A difficult opening should not swallow the paper

Some texts begin densely and become clearer once the topic, speakers or definitions settle. If a learner treats the first paragraph as a gate that must be perfectly solved before continuing, the opening can consume disproportionate time.

Use the anchor method: extract the stable proposition, mark unresolved detail and continue far enough to see whether later context repairs it. Return if the unresolved point proves structurally important.

Leonie practises this with passages deliberately chosen for opaque openings. Her success criterion is not instant certainty. It is maintaining a viable model despite an imperfect first minute.

86. A difficult middle often signals a local compression problem

Writers may compress several claims, sources or causal steps into one middle paragraph. Readers who were comfortable in the opening can suddenly feel the whole passage has become difficult.

Localise. What new job does the paragraph perform? Which proposition is central? Are multiple sources being compared? Is a long sentence carrying claim, evidence and qualification at once? Break the section into propositions, then reconnect it to the larger map.

The key is to avoid letting one dense region retroactively erase understanding of the rest of the text.

87. A difficult ending should be read against the established model

Conclusions can be abstract because they compress the implications of everything before them. Instead of reading the ending in isolation, ask what earlier material it is returning to, qualifying or resolving.

If the final sentence seems surprising, compare it with the opening claim and major turn. Does it narrow the argument? Leave a question open? Recast a repeated image? Move from evidence to recommendation?

A model built through the passage becomes a tool for understanding the ending. The ending should update the model, not require an entirely new one without evidence.

88. Reading under pressure needs a recovery hierarchy

When meaning breaks, use the smallest repair first. Check reference. Check relationship words. Find the grammatical spine. Revisit the previous stable proposition. Widen to the paragraph job. Only then consider rereading a larger section.

This hierarchy prevents panic from triggering the most expensive response—starting again from the top. It also gives the learner something concrete to do when a sentence feels impossible.

Iona’s repair order becomes automatic with practice: local reference before global reread. The hierarchy is efficient because many comprehension failures begin with one broken connection, not total loss of meaning.

89. Anxiety changes what the reader notices

Under pressure, difficult vocabulary appears larger, ambiguity feels dangerous and the urge to underline increases. The reader may spend more time proving obvious details and less time tracking relationships. Anxiety changes attention allocation.

A stable unseen routine protects against this shift because it tells the learner what to prioritise: object, viewpoint, movement, turn, uncertainty, map. The routine does not remove emotion. It gives attention somewhere productive to go.

Leonie practises with gradually increasing time pressure so the routine is not learned only in calm conditions. The system must survive the state in which it will be used.

90. Annotation should shrink as expertise grows

Beginners may need explicit paragraph labels, reference arrows and uncertainty marks because the relationships are not yet easy to hold mentally. As recognition improves, excessive annotation becomes unnecessary overhead.

Fade supports deliberately. First require labels. Then allow only one or two words per paragraph. Later ask the learner to mark only turns and high-risk uncertainties. Compare performance. If accuracy remains stable while time improves, the representation is becoming internal.

Expert reading often looks less busy on the page because more of the structure is being carried mentally.

91. Do not confuse a neat page with a strong model

Colour-coded annotation can look impressive while hiding weak comprehension. A learner may identify every metaphor and connector yet be unable to state the writer’s final position. Another may make only three marks and answer accurately because those marks represent decisive shifts.

Evaluate annotation by retrieval and answer performance, not appearance. Can the learner use the markings to locate evidence? Do they preserve the passage movement? Are they faster or slower because of the system?

The examination rewards understanding expressed in answers. The margin is a tool, not a product.

92. Reading notes should use verbs

Functional notes become more useful when they contain action: challenges claim, narrows conclusion, introduces cause, returns to memory, shifts audience, concedes cost. Verbs encode movement.

Noun-only notes—technology, problem, childhood, statistics—record topics but not relationships. When the learner later searches for evidence, topic notes may identify several paragraphs without distinguishing their jobs.

Maren develops a simple habit: every paragraph label must answer “What does this section do?” rather than “What is it about?” The change improves navigation immediately.

93. Use arrows sparingly for genuine relations

Arrows can store cause, consequence, contrast, reference or development. They become useless when everything is connected to everything. The reader should use them only when the relation is important enough to retrieve later.

For example, “policy delay → cost increase” may preserve a causal link across paragraphs. “claim ↔ counterclaim” may mark a structural opposition. An arrow from every sentence to the next merely redraws the passage.

The best annotation system has low symbol vocabulary and high interpretive value.

94. Margin questions can preserve uncertainty without derailing reading

A tiny question—“why now?”, “whose view?”, “literal?”—can hold an unresolved issue until later context arrives. This is useful when the question matters enough not to forget but not enough to stop the first read.

Return after the passage. Some questions will have answered themselves. Others may align with the examination tasks. A few may remain genuinely ambiguous, in which case the learner knows where interpretation requires caution.

This technique converts uncertainty from a blockage into a tracked state.

95. The reader should know when not to annotate

If a section is clear, structurally ordinary and unlikely to be forgotten, annotation may add nothing. Continue reading. Every pause breaks flow slightly, and flow matters when building global meaning.

Leonie’s most important annotation lesson is subtraction. She once marked every unfamiliar word. Now she marks only central unknowns, major turns and relationships that would otherwise require rereading. Her page looks quieter and her later search becomes faster.

Selective silence in the margin is part of reading discipline.

96. Narrative maps should preserve pressure, change and consequence

A useful narrative map is not a plot summary. It preserves the relationships that later questions are likely to test: starting situation, pressure, response, change and consequence. If the learner can remember only what happened but not what changed because of it, the map is too flat.

For a short extract, Maren may compress the movement into four verbs: waits → notices → hesitates → acts. Iona then adds one controlled interpretive layer: “hesitation increases after the letter is mentioned.” The map remains grounded in observable events while becoming useful for inference. Leonie adds location only when it matters for later retrieval.

The test is simple: could this map help you find the moment when the character’s attitude changes without rereading the whole passage? If yes, it is doing examination work.

97. Argument maps should preserve claim, support, challenge and final position

Argumentative unseen texts become easier when the reader stops treating every paragraph as another opinion. Map the hierarchy: central claim, main reasons, evidence, counterargument, concession and final position. The writer may not follow this order neatly, but these functions remain useful.

Record the final claim separately from the opening claim. A sophisticated article may begin strongly, concede a limitation, then end with a narrower recommendation. If the map stores only the first position, later questions about stance will be answered from an outdated model.

Iona’s shorthand might be: “benefit claimed → study supports → cost challenge → concession → support retained, but only at small scale.” That one line contains enough architecture to predict where several kinds of evidence will sit.

98. Explanatory maps should preserve mechanism, conditions and limits

Explanatory prose often answers a hidden question: how does this work, why does this happen, or under what conditions does this occur? A good map therefore preserves the mechanism rather than merely listing facts.

Suppose a passage explains urban heat. The map should not become “roads, buildings, trees, night temperatures.” A stronger map is: “dark surfaces absorb heat → sparse vegetation reduces cooling → dense structures slow heat loss → effect strongest at night.” The arrows preserve causality and sequence.

Then attach limits: “effect varies with weather and design.” Later questions about causes, exceptions or evidence can now be answered from a structured model rather than a bag of technical nouns.

99. Biography and memoir need a timeline plus an interpretation boundary

Biographical and memoir extracts mix event with retrospective meaning. The narrator may describe what happened years earlier while interpreting it from a later position. The map should preserve both the event timeline and the reflective layer.

Mark when the text moves from “what happened then” to “what I understand now”. A remembered embarrassment may be narrated humorously in the present. A childhood decision may acquire significance only because the older narrator knows its consequences. The current interpretation should not be projected automatically onto the younger self.

This distinction gives later questions about tone, hindsight and change a clean evidential base.

100. Speeches and letters require an audience map

In speeches, open letters and public statements, audience is structural. Ask who is being addressed, what shared ground is assumed, what resistance the writer expects and what response the text ultimately seeks.

The audience map may change. A speaker can begin with inclusive “we”, then challenge one subgroup, then return to common purpose. A letter may move from personal gratitude to public criticism. Direct address can create intimacy, pressure or responsibility depending on context.

Maren maps audience movement as carefully as topic movement because persuasive effect often depends on how the relationship with the reader is managed over time.

101. Editorials and commentary need stance plus evidence hierarchy

An editorial can sound confident while using evidence of very different quality. Separate stance from support. What does the writer want accepted? Which evidence is empirical, which is anecdotal, which comes from authority, and which is interpretation?

The map should also preserve concessions and rhetorical framing. A strong phrase may reveal attitude without adding evidence. A statistic may support one narrow point without proving the larger conclusion. Keeping those functions separate makes later evaluation more disciplined.

Iona’s first-pass note can be compact: “strong stance; mixed support; accepts implementation cost; rejects delay.” That is enough to guide a later close read.

102. Science and technical texts need proposition discipline

Technical vocabulary can intimidate readers into abandoning structure. Yet scientific explanations still contain claims, evidence, conditions, comparisons and uncertainty. The reader does not need full subject expertise to recognise those relations.

Track what was observed, what was inferred and how certain the writer is. “Researchers observed X” differs from “researchers propose Y as an explanation.” “The results are consistent with” is weaker than “the results demonstrate.” These distinctions often matter more than memorising every specialised term.

A good map lets the learner say, “study finds association; mechanism proposed; limitation acknowledged.” That is already enough architecture for many comprehension questions.

103. Historical accounts need chronology plus source position

Historical prose often combines sequence, explanation and interpretation. Keep chronology stable first: what happened before what? Then track the writer’s causal claims and source position. A later historian’s explanation is not the same as what participants at the time believed.

Watch for phrases such as “at the time”, “later evidence suggests”, “contemporaries believed” and “historians have argued”. These mark changes in knowledge state. The passage may contrast contemporary perception with retrospective analysis.

Leonie uses a simple two-line map when needed: event sequence on one line, interpretations on the other. That prevents explanation from becoming confused with chronology.

104. Paired-text maps should remain separate until the comparison question arrives

For two texts, create two compact maps rather than one blended note. Each map should contain position, development, evidence and final stance. Only after both are stable should the reader build comparison axes.

This protects against confirmation bias. If Text A is clearly enthusiastic, the reader may otherwise hunt Text B for opposition and overlook a more nuanced difference such as degree, evidence base or audience. Separate maps let the comparison emerge from the texts rather than from an imposed binary.

The rule is simple: understand separately, compare relationally.

105. Multimodal maps should show how modes cooperate

When prose, image, graph and caption appear together, the map should record their relationships rather than create four separate summaries. Does the image humanise the issue the statistics quantify? Does the graph challenge the headline? Does the caption direct interpretation of an otherwise ambiguous photograph?

The strongest question is: what does this mode contribute that the others do not? That answer preserves function and prevents visual material from becoming decorative in the learner’s mind.

Iona may write: “prose claims improvement; chart shows uneven gains; caption narrows time period.” The relationship is now visible before any question asks for it.

106. Train paragraph-function recognition before full-paper speed

If learners cannot identify what paragraphs do when untimed, timing them will not create the skill. Use short passages and ask for one functional verb per paragraph: defines, challenges, exemplifies, concedes, explains, contrasts, narrows, concludes.

Then remove the labels and ask learners to predict where a hypothetical question would send them. “Where would you look for the limitation?” “Which paragraph contains the reason the writer changes position?” This connects structure directly to later retrieval.

Only after function recognition becomes reliable should the exercise be compressed under examination timing.

107. Train turn detection with before-and-after statements

Give learners passages containing a meaningful shift and ask them to write two sentences: “Before the turn…” and “After the turn…”. This forces them to identify both sides of the movement rather than merely circle a connector.

Later remove obvious transition words. Use gradual changes in tone, evidence or character behaviour. Ask where the model must be updated even though no single sentence announces the shift.

The purpose is not to make every passage hinge on one dramatic pivot. It is to make readers sensitive to development, because development is one of the main reasons a passage needs to be read as a whole.

108. Train reference repair as an emergency skill

Reference chains are ideal for small diagnostic drills. Give a paragraph with several pronouns and compressed phrases. Ask learners to replace each uncertain reference with its antecedent and explain why competing candidates fail.

Then place the same skill back inside a longer unseen passage where only one reference matters. The learner should recognise when to deploy the repair rather than tracing every pronoun automatically.

This matters because reference failure is often a hidden first weak link. Later inference may look sophisticated while being built on the wrong subject.

109. Train uncertainty tolerance deliberately

Give learners passages containing one unresolved word, ambiguous motive or temporarily unclear reference. Ask them to continue reading while marking the uncertainty lightly. Then review whether later context resolved it and whether immediate stopping would have helped.

The goal is calibrated tolerance. Some uncertainty can be carried. Structural uncertainty cannot. A learner should know the difference between “I do not know this adjective exactly” and “I no longer know who is speaking.”

Leonie improves when she learns that temporary incompleteness is not failure. It is a normal state in unseen reading.

110. Train model updating, not first-guess loyalty

Use passages that deliberately complicate an early interpretation. Ask learners to state their provisional model after the opening, then revise it after new evidence. The score should reward justified updating rather than punishing the initial provisional hypothesis.

This builds intellectual flexibility. Strong readers are not those whose first guess is always right. They are readers who detect mismatch, locate the evidence causing it and revise without distorting the text to protect an earlier idea.

Iona records the change explicitly during practice: “first thought → new evidence → revised model.” Over time, the update becomes mental and fast.

111. Train with familiar topics to expose assumption errors

Familiar content is a valuable stress test because knowledge can mask weak reading. Give learners passages on subjects they know well and ask them to separate what the passage states from what they already believe. When the source uses a narrower definition, unusual example or qualified conclusion, the model must follow the source.

Review any answer that contains information absent from the text. Was the extra knowledge useful context, or did it silently replace the writer’s proposition? This exercise reveals a sophisticated failure mode: the learner is not confused by the text; the learner is too confident about the topic.

True transfer works both when background knowledge is thin and when it is abundant.

112. Remove headings to test whether the learner can recover structure

Headings provide free architecture. Once learners can use them well, remove them from selected practice passages and ask the learner to reconstruct the missing section labels. What job would a useful heading need to perform here?

This exercise reveals whether the learner was reading the prose or simply following supplied signposts. It also improves paragraph-function recognition because the learner must identify where one unit ends and another begins.

Later restore headings and compare. Did the writer’s actual labels emphasise the same structure the learner inferred? Differences can reveal framing and audience choices.

113. Increase syntax density independently of topic difficulty

To diagnose sentence processing, use passages on accessible topics written with more complex syntax. If comprehension falls sharply while vocabulary remains familiar, sentence architecture is likely contributing to the difficulty.

Train learners to identify main clauses, embedded qualifications and logical connectors without rewriting every sentence. Then return to the passage map. The purpose of syntax repair is not to turn reading into grammar analysis; it is to restore proposition structure quickly enough that the global model survives.

As control improves, combine syntax density with unfamiliar topics so the reader must use both structural and contextual strategies.

114. Increase paraphrase distance between the text and later questions

A strong passage model should survive when later questions use different vocabulary. In practice, gradually increase the semantic distance between source and task. A paragraph about “declining public confidence” may later be queried through wording about “loss of trust”.

If the learner can navigate only when the question repeats visible words from the passage, the map is lexical rather than conceptual. Ask the learner to locate where the idea lives without searching for identical wording.

This training belongs at the boundary between passage modelling and evidence search: the map supplies conceptual locations; the next lane skills turn those locations into exact corridors.

115. Increase viewpoint complexity one layer at a time

Begin with texts where writer and speaker positions are easy to distinguish. Then add quoted voices, unreliable narrators, reported beliefs and competing sources. Complexity should rise while attribution discipline remains stable.

A useful exercise asks learners to tag each important proposition by owner: narrator, character, researcher, critic, writer, source A, source B. Then remove the tags and test whether the learner can maintain the separation mentally.

Viewpoint errors are costly because they can contaminate multiple later answers. Training attribution early creates a cleaner foundation for inference and evaluation.

116. Introduce multiple voices without making every voice equal

A passage may quote five people while still having one controlling writer. Mapping every voice with equal weight can make the text seem more fragmented than it is. Ask which voices are central, which provide examples, which are challenged and which appear only briefly.

Hierarchy matters. A one-line quotation from a resident may humanise a problem; a research report may supply the main evidence; the writer may use both to support a conclusion. The map should reflect that architecture rather than count speakers.

Maren’s shorthand is functional: “research = proof; resident = human impact; official = counterclaim.” Later questions can then retrieve the right voice quickly.

117. Test whether the passage map survives late-paper fatigue

A learner may build excellent maps early in practice and poor ones after forty minutes. Fatigue can flatten paragraph functions into topics, increase rereading and reduce tolerance for uncertainty. Therefore, test passage modelling at realistic positions inside longer sessions.

Compare early and late maps. Does the learner stop identifying turns? Do annotations become heavier? Are quantifiers forgotten? Does the final position get confused with an earlier claim? These are procedural fatigue signals.

The repair may be a shorter first-pass routine, better pacing earlier in the paper or targeted endurance work. “Concentrate harder” is not a diagnosis.

118. Calibrate confidence in the passage map itself

After a first read, ask the learner to mark which parts of the model are high, medium or low confidence. The purpose is not to create hesitation. It is to distinguish secure structure from provisional interpretation.

Iona may be highly confident that paragraph four introduces a counterargument but only moderately confident about the writer’s tone toward it. That distinction tells her what needs close rereading if a later question targets the section.

Over time, compare confidence with accuracy. A learner who is repeatedly overconfident about narrator motive needs different training from one who distrusts accurate structural readings.

119. Keep a passage-modelling error ledger

Not every comprehension error belongs in one undifferentiated list. For first-read modelling, useful categories include lost reference, missed turn, wrong voice attribution, topic-not-function mapping, scope loss, premature interpretation, cause-sequence confusion, unknown-word fixation, annotation overload and failure to update.

Record the first modelling error that made later work harder. If a wrong inference began because the reader attached “they” to the wrong group, the ledger should record reference failure rather than inference failure.

Patterns across several unseen texts reveal which part of the first-read system deserves training priority.

120. Correct, contrast and transfer the passage map

After a weak first read, do more than show the teacher’s summary. First reconstruct the corrected map. Then contrast it with the learner’s map: where did they diverge, and what signal was missed? Finally, transfer the distinction to a fresh text.

If the learner missed a concession, compare a passage where “although” preserves the main claim with one where a later sentence genuinely reverses it. If the learner blended sources, use a new multi-voice text and require attribution before interpretation.

Correction teaches the original text. Contrast teaches the boundary. Transfer tests whether the boundary was learned.

121. A one-week unseen-reading cycle

Day 1 establishes a baseline with two unfamiliar passages and records map quality, rereading and time. Day 2 isolates paragraph jobs and turns. Day 3 trains reference, scope words and model updating. Day 4 uses contrasting text families—one narrative, one argument. Day 5 adds moderate timing. Day 6 retests the dominant error on new material. Day 7 uses one short transfer passage and light retrieval of the operating routine.

The cycle should not become ceremonial. If the learner’s main weakness is technical vocabulary fixation, allocate more practice to uncertainty triage. If structure is strong but timing remains poor, reduce annotation and audit rereading.

The plan follows evidence.

122. A six-week build for unseen passage modelling

Week 1 diagnoses map quality across several text types. Week 2 stabilises paragraph function, reference and logical relations. Week 3 develops narrative and argumentative maps. Week 4 adds multiple voices, multimodal material and wider subject domains. Week 5 compresses the routine under realistic timing and practises recovery from deliberately difficult sections. Week 6 uses full comprehension sections and delayed transfer tests.

Broad vocabulary growth continues throughout, but the six-week goal is not to make every word familiar. It is to make first-read decision quality more robust when unfamiliarity appears.

Each week should revisit one older distinction so gains are tested for retention rather than immediate recognition.

123. A twelve-week build moves from representation to automaticity

Weeks 1–2 establish the diagnostic baseline and explicit mapping routines. Weeks 3–4 develop structure, viewpoint and uncertainty control. Weeks 5–6 broaden text families and domain knowledge. Weeks 7–8 introduce paired and multimodal texts. Weeks 9–10 emphasise timing, reduced annotation and fatigue resistance. Week 11 uses realistic simulations. Week 12 compresses the routine to the smallest set of decisions the learner still needs consciously.

The long build allows supports to fade. What begins as written paragraph labels should become mental recognition. What begins as explicit confidence tagging should become calibrated judgement. What begins as a repair hierarchy should become automatic recovery.

The destination is not more visible method. It is reliable invisible method.

124. Teacher diagnosis should compare maps before answers

Before reviewing question marks, ask learners to reconstruct what they thought the passage was doing. Two students can choose the same wrong answer because one lost the writer’s position and the other understood the passage but misread the question. Their repairs are different.

Collect compact passage maps occasionally. Look for paragraph functions, major shifts, source attribution and certainty control. Compare the maps with later answer patterns. A student who consistently builds accurate maps but loses marks during answer construction should not spend weeks reteaching first-read strategy.

Diagnosis becomes sharper when each stage of the comprehension chain can be observed separately.

125. Self-study should make the hidden map visible

Independent learners can train passage modelling by writing a five-line reconstruction after each practice text: text type, controlling viewpoint, section movement, major turn, final position or state. Then compare it with the questions and marked answers.

If a question repeatedly sends the learner to a location not represented in the map, ask what function was missed. If the map is accurate but evidence search remains slow, the next repair belongs to a later stage rather than to unseen reading itself.

Self-study improves when the learner stops measuring only correctness and begins measuring the quality of the internal representation that produced it.

Technical Appendix A: Passage-map templates by text family

Templates are useful only if they remain provisional. Their purpose is to reduce the cost of orientation, not to force every unseen passage into a predetermined shape. A template tells the learner what relationships are worth looking for; the text decides whether those relationships are actually present.

A1. Narrative and literary extract

• Starting state: who is present, where, and what appears normal?
• Pressure: what new event, observation, memory, person or conflict changes the state?
• Viewpoint: through whose perception is the scene filtered?
• Development: what changes in behaviour, knowledge, relationship, pace or attention?
• Turn: where does expectation, mood or understanding shift?
• Ending state: what is different by the end, and what remains unresolved?

A concise map might read: “quiet wait → stranger arrives → narrator becomes watchful → conversation exposes old connection → narrator reinterprets earlier silence.” This is not a plot summary because each arrow describes a change relevant to later questions.

A2. Argument and commentary

• Issue: what question or problem is being disputed?
• Opening position: what does the writer initially appear to claim?
• Support: which reasons, data, examples or authorities carry the claim?
• Pressure: what objection, limitation or competing viewpoint enters?
• Response: does the writer reject, absorb or concede the challenge?
• Final position: is the conclusion stronger, narrower or differently framed?

Do not let a memorable anecdote replace the central claim. Do not let a concession replace the final position. The template exists to preserve hierarchy.

A3. Explanation and science communication

• Phenomenon: what needs explaining?
• Mechanism: what sequence or relationship makes it happen?
• Evidence: what observation or study supports the mechanism?
• Condition: when does the explanation apply?
• Limit: what uncertainty or exception remains?
• Implication: why does the mechanism matter?

The mechanism should be represented with verbs and relations rather than a noun list. “Warm water rises → cooler water replaces it → circulation distributes heat” is more useful than “water, heat, circulation”.

A4. Memoir and biography

• Then: what happened in the original time frame?
• Then-self: what did the person know or believe at that time?
• Now: from what later perspective is the event being interpreted?
• Reframing: how does hindsight change the meaning of the event?
• Significance: what larger idea does the remembered event now support?

This two-time-layer map prevents hindsight from being attributed to the younger self and helps the learner recognise reflective structure.

A5. Paired or multiple sources

Map each source independently using four fields: position, development, evidence, qualification. Only then create comparison axes. A useful comparison map may look like: “A supports expansion because benefits appear scalable; B supports limited trials because implementation evidence remains uneven.” The axis is degree of support, not simply positive versus negative.

Technical Appendix B: Diagnostic matrix for unseen reading

When a learner says, “I did not understand the passage,” the statement is too broad to guide repair. Use observable symptoms to infer the likely first weak link, then test the hypothesis with a smaller task.

B1. The learner remembers details but cannot explain the passage movement

Likely cause: reading at sentence level without paragraph-function integration. Smallest repair: one functional verb per paragraph, followed by a five-line whole-text map. Transfer test: a new passage where topic vocabulary differs but structural movement is similar.

B2. The learner misattributes a quotation or opinion

Likely cause: unstable voice map. Smallest repair: label propositions by source or speaker, then explain what the controlling writer does with each voice. Transfer test: a new article containing at least three reported positions.

B3. The learner rereads from the top after every difficult sentence

Likely cause: no recovery hierarchy or low uncertainty tolerance. Smallest repair: require local repair in this order—reference, connector, main clause, previous stable proposition—before broad rereading is permitted. Transfer test: insert one deliberately dense sentence into an otherwise accessible passage and record the learner’s response.

B4. The learner understands familiar topics but collapses on unfamiliar ones

Likely cause: background knowledge is carrying the reading. Smallest repair: use unfamiliar passages with clear discourse structure and ask for relationships before detailed content. Transfer test: a second unfamiliar domain with different vocabulary but similar logical architecture.

B5. The learner overinterprets narrative details

Likely cause: early labels becoming fixed conclusions. Smallest repair: record behaviour first, interpretation second, and require more than one clue for a strong character claim. Transfer test: a passage where the apparent early motive is later revised.

B6. The learner underlines heavily but searches slowly later

Likely cause: annotation stores salience rather than function. Smallest repair: limit marks to paragraph job, turn and high-risk uncertainty. Transfer test: compare answer time and accuracy across two equally difficult passages using old and reduced annotation systems.

B7. The learner misses qualification

Likely cause: topic compression discards modality, quantifiers or conditions. Smallest repair: rewrite main claims while preserving must/may, some/all, only/unless and similar scope controls. Transfer test: compare two passages whose topics match but certainty differs.

Technical Appendix C: Three worked passage-model scenes

C1. Iona and the misleading first paragraph

Iona reads an article that opens by celebrating a city’s new bicycle network. The language is energetic: more lanes, more riders, fewer short car journeys. Her first hypothesis is “writer supports expansion.” In paragraph two, the writer introduces data showing strong use in the centre but little access in outer districts. Paragraph three quotes residents who describe unsafe links between neighbourhoods. The conclusion supports further expansion only if the network becomes connected rather than merely larger.

A weak first read stores “positive article about cycling”. A stronger map updates: “opening success → distribution problem → resident evidence → final support becomes conditional on connectivity.” The writer did not reverse position completely. The position became narrower and more demanding. If a later question asks about attitude, the map already protects Iona from calling the writer simply enthusiastic or opposed.

The lesson is model revision. The first paragraph was not wrong; it was incomplete. Later evidence changed the scope of the conclusion.

C2. Leonie and the difficult technical sentence

Leonie meets a science passage containing a long sentence about urban trees, evapotranspiration and surface temperatures. One term is unfamiliar and the clause structure is dense. Her old habit is to restart the paragraph twice. Her new routine is local. She finds the main clause: “Trees can reduce local temperatures.” She identifies two supporting mechanisms: shade and water loss from leaves. The unfamiliar term appears to name the second mechanism more technically.

She continues. The next paragraph introduces a limitation: cooling depends on species, water availability and street design. Her map becomes “trees cool through shade + evaporation → effect varies by conditions.” She has preserved the passage architecture without needing dictionary-level precision on first contact.

The lesson is localisation. One unknown technical word did not require surrendering the paragraph or the clock.

C3. Maren and the literary symbol that might not be a symbol

Maren reads a story in which a character repeatedly checks a train timetable. She immediately wants to interpret the timetable as a symbol of escape. Instead, she records the concrete pattern. At first the character checks it while waiting for a friend. Later she checks it after an argument. Near the end she folds it and puts it away before choosing to walk home.

Now an interpretation is better supported. The timetable is connected to possible departure, but its meaning develops through the character’s decisions. The final act of putting it away may suggest refusal, delay or a change in intention; the exact interpretation depends on surrounding details. Maren’s passage map stores the transformation before it stores the symbolic label.

The lesson is evidential patience. The reader lets pattern create interpretation instead of forcing interpretation onto the first occurrence.

Technical Appendix D: Stress tests for transfer

Training should deliberately make one dimension harder at a time. This reveals whether the unseen-reading system is genuinely portable or dependent on comfortable surface conditions.

1. Vocabulary stress: use an unfamiliar domain but keep syntax and structure clear. Can the learner preserve movement while tolerating lexical gaps?
2. Syntax stress: use familiar content with dense sentences. Can the learner recover the grammatical spine without losing the whole paragraph?
3. Voice stress: include several quoted speakers. Can the learner maintain attribution and identify the controlling writer?
4. Turn stress: remove obvious connectors and let the stance change gradually. Can the learner recognise model revision from distributed evidence?
5. Familiarity stress: choose a topic the learner knows well but write a passage that takes an unexpected position. Can the learner obey the text rather than prior belief?
6. Multimodal stress: combine prose with a graph whose data qualifies the headline claim. Can the learner model the relationship across modes?
7. Fatigue stress: place the unseen passage late in a realistic session. Does paragraph-function reading survive, or does the learner fall back to topic notes and broad rereading?
8. Time stress: compress the first-read window gradually. Which part of the map disappears first, and can training stabilise it?

A stress test should answer one diagnostic question. Do not increase vocabulary, syntax, voice complexity, time pressure and fatigue simultaneously unless the purpose is final simulation. Controlled difficulty makes causes visible.

Technical Appendix E: The final unseen-text operating routine

By examination day, the routine should be compact enough to execute without narrating every step internally. The following expanded version is useful during training; the mature version should compress into fluent behaviour.

1. Orient. Identify text family, available context, viewpoint and likely purpose without forcing a final interpretation.
2. Read for movement. Track what changes from section to section rather than collecting equal quantities of detail.
3. Protect relations. Notice cause, contrast, concession, condition, reference, sequence, modality and scope words that control propositions.
4. Build paragraph jobs. Know what major sections do: introduce, explain, exemplify, challenge, concede, shift, resolve.
5. Hold uncertainty. Carry non-structural unknowns; repair structural breaks locally before broad rereading.
6. Update. Revise the model when later evidence changes viewpoint, certainty, scale or interpretation.
7. Find the turn. Identify major movement in stance, mood, knowledge, time or relationship where present.
8. Finish the map. Retain the starting state, major stages, final position and high-value unresolved questions.
9. Prepare return points. Know roughly where causes, examples, challenges, changes and conclusions live.
10. Then answer. Shift from whole-text modelling to the question-to-mark chain rather than continuing to read the passage indefinitely.

The first read is complete when the text has become navigable, not when every sentence feels exhausted. That distinction is the heart of unseen-text performance. A learner who demands perfect understanding before answering will always be vulnerable to difficult material. A learner who can construct a reliable provisional model can continue working while uncertainty remains.

126. How SETC-SEC-COMP-020 connects to the rest of the lane

This page covers unseen-text passage modelling: orientation, first-read architecture, viewpoint separation, structural mapping, uncertainty control and preparation for later retrieval. It does not cover the complete mechanics of inference, vocabulary-in-context or answer precision. Those remain routed through the existing reading foundation and related performance guides.

Return to SETC-SEC-COMP-000 for the full Secondary comprehension performance system. Use SETC-SEC-COMP-010 when the problem begins after the question appears. Use English Reading Performance for the broader foundation in evidence, inference and answer scope.

The next lane problem is time. A strong map can still fail if the learner spends too many minutes building it, allocates time equally to unequal questions or lets one difficult item consume the paper. That job belongs to SETC-SEC-COMP-030: Reading Comprehension Time Management.

127. Final principle: make the unfamiliar navigable

The purpose of unseen-text training is not to make future passages predictable. It is to make the learner’s response to unpredictability reliable.

Iona learns to update without over-inference. Leonie learns to localise difficulty without sacrificing the clock. Maren learns to see architecture without turning every detail into symbolism. Together they illustrate the mature unseen reader: attentive but selective, interpretive but evidence-bound, flexible but not vague.

YOU DO NOT NEED TO KNOW THE TEXT BEFORE THE EXAM. YOU NEED TO KNOW HOW TO BUILD A WORKING MODEL WHEN THE TEXT ARRIVES.

That is the portable advantage. The passage changes. The method survives.

128. Practice Lab: build a map from a passage that changes its mind

Consider a fictional article about remote work. The opening presents lower commuting time, quieter offices and wider recruitment as clear benefits. A middle section then reports that new employees learn informal routines more slowly when contact is mostly scheduled. Another paragraph describes teams that solved the problem through structured mentoring and regular in-person project days. The conclusion rejects a simple choice between “office” and “remote” and argues for different arrangements according to the kind of work being done.

A weak map says, “advantages and disadvantages of remote work.” That topic label loses the movement. A stronger map is: “opening celebrates efficiency → onboarding evidence introduces hidden cost → hybrid practices show cost can be reduced → conclusion replaces universal rule with task-dependent model.” The final position is not neutral indecision. It is a new criterion for deciding.

Now ask what a later question might target. A tone question could ask why the writer becomes more qualified. A structure question could ask why mentoring examples appear after the problem. An evaluation question could ask whether the evidence justifies the final recommendation. The passage map has already predicted the likely return points without answering those questions prematurely.

129. Practice Lab: preserve a narrative before interpreting it

Imagine an extract in which a teenager arrives early at a nearly empty sports hall. She checks the scoreboard although no game has begun, straightens three chairs, reads the same message on her phone twice and puts the phone face down when footsteps approach. The arriving person is her former teammate, whom she greets with an ordinary question about the weather.

An impatient reader labels her anxious immediately. A disciplined first map records the observable pattern: “arrives early → repetitive preparation/checking → hides or ends phone attention when footsteps arrive → conversation begins indirectly.” Anxiety is plausible, but so are anticipation, awkwardness and deliberate self-control. The former teammate’s arrival makes the repeated actions more meaningful, yet motive remains partly open.

If a later paragraph reveals that the teenager had asked to meet in order to apologise, the model updates. The early actions can now be interpreted as preparation under social pressure. The reader has lost nothing by delaying the label; instead, the later evidence produces a stronger interpretation.

130. Practice Lab: distinguish a claim from the evidence surrounding it

Imagine a passage arguing that city parks improve neighbourhood resilience during heatwaves. The writer opens with temperature measurements from several districts, then tells the story of an elderly resident who walks to a shaded park during hot afternoons. A researcher explains how tree canopy changes surface temperature. The writer later admits that parks alone cannot protect residents without suitable housing and public-health planning.

The map should preserve evidence hierarchy. Measurements establish a temperature pattern. The resident story makes human consequence concrete. The researcher explains a mechanism. The concession prevents the claim from becoming “parks solve heat risk”. A strong map is: “parks associated with cooler local conditions → personal case humanises benefit → mechanism explains cooling → concession places parks inside wider resilience system.”

Because the map separates function, a learner can later evaluate each piece of evidence without confusing vividness with generalisability.

131. Practice Lab: map two sources before comparing them

Source A describes artificial intelligence in classrooms as useful when teachers design the task, check outputs and require students to explain decisions. It uses examples of feedback and language practice but warns against unverified answers. Source B focuses on assessment integrity. It argues that unsupervised use can hide what a learner can actually do, yet accepts AI for practice outside formal assessment.

If a learner begins with “A supports AI, B opposes AI”, the comparison has already distorted both texts. Separate maps are better. A: “supports guided learning use → examples of feedback → verification condition.” B: “restricts assessment use → concern about evidence of learning → accepts practice use.” Only after those maps are stable should the comparison axis emerge.

The more accurate synthesis is that both sources accept some educational use but draw the safety boundary in different places. Source A focuses on instructional control; Source B focuses on validity of assessment. That distinction could not emerge from a binary positive-versus-negative model.

132. Practice Lab: recover from a dense sentence without restarting

Suppose a sentence states: “Although the early trials, conducted in relatively stable conditions and involving volunteers already familiar with the technology, produced encouraging results, later deployments across less predictable environments exposed limitations that the original evaluation had not been designed to measure.”

The sentence looks long because the main contrast is interrupted by qualification. Strip it to the spine: “Although early trials produced encouraging results, later deployments exposed limitations.” Then reattach the information: early trials were stable, involved experienced volunteers, and did not measure the later limitations. The sentence is no longer a wall; it is a contrast plus an explanation of why the original evidence may not generalise.

Leonie should not restart the entire paragraph after solving this sentence. She updates the map: “promising controlled trials → real-world conditions reveal measurement gap.” Local repair restores global movement.

133. Practice Lab: decide which unknown word deserves attention

Imagine a passage saying that a committee’s initial response was “perfunctory”, followed by a description: members approved the proposal after only a few minutes, asked no questions and moved immediately to the next agenda item. A learner may not know “perfunctory”. The surrounding evidence nevertheless establishes a rough meaning: brief, routine, lacking careful attention.

Now compare a second unknown word—the technical name of a device mentioned in an example but irrelevant to the paragraph’s main relationship. That word can probably be postponed. The first unknown describes the committee’s behaviour and may matter for tone or attitude. The second names an object the passage itself does not develop.

Vocabulary triage asks not merely “Do I know this word?” but “What decision depends on knowing it more precisely?” That question protects both comprehension and time.

134. Practice Lab: detect a quiet structural turn

Not every turn begins with “however”. Consider a passage that spends three paragraphs explaining why consumers prefer convenience. The fourth paragraph begins, “The preference looks different when the cost is delayed.” No dramatic opposition marker appears, but the criterion has changed. The writer moves from immediate experience to long-term consequence.

A topic map may miss the turn because every paragraph still discusses consumer choice. A functional map catches it: “explains convenience preference → changes time horizon → reveals hidden cost → argues for different decision standard.” The text has not changed subject; it has changed the lens through which the subject is judged.

Training quiet turns prepares learners for sophisticated prose where development is carried by concepts rather than obvious connectors.

135. Practice Lab: prevent a graph from overpowering the prose

Suppose an article headline claims that public transport satisfaction has “surged”. A chart shows satisfaction rising from 68 to 73 per cent over four years. The prose then explains that most of the improvement occurred after service frequency increased on two heavily used routes, while other areas changed little.

The graph supports improvement but the word “surged” may frame the size more dramatically than the numbers alone justify. The later prose also makes the improvement uneven. A good multimodal map becomes: “overall rise modest → concentrated on improved routes → headline uses stronger framing than distribution suggests.”

The map does not accuse the writer of deception. It distinguishes data, distribution and framing so that a later question about evidence or effect can be answered proportionately.

136. Practice Lab: preserve qualification in memory

Read these two propositions: “The intervention improved outcomes for most participants when weekly support continued,” and “The intervention improved outcomes.” The shorter version is easier to remember but materially broader. It has removed both the quantifier and the condition.

During unseen reading, compression is necessary; distortion is not. The learner can store the claim efficiently as “helps most, with weekly support.” That compact representation preserves what matters. If the support condition disappears from the passage map, later questions about effectiveness may be answered too strongly.

Practise compressing ten qualified claims into short notes while retaining quantity, certainty and condition. This develops a form of semantic shorthand more useful than copying whole sentences.

137. Practice Lab: recognise when the model is good enough

Imagine Iona finishes a first read and can state the following: “The article begins by presenting a technology as convenient, then introduces privacy concerns, then shows two safeguards, and ends by supporting use only with stronger consent rules. The regulator’s quotation is the main counterweight to the company. I am unsure about one technical term in paragraph three, but it does not seem to change the argument.”

That is a functional map. Iona does not need to reread the entire article merely because one term remains unclear. She knows the controlling movement, source roles, major turn and final position. The text is navigable.

The discipline of stopping matters as much as the discipline of reading. An unseen first pass should end when it has created enough structure to support efficient question work. Perfection is not the target; recoverable orientation is.

Technical Appendix F: 30-second post-read reconstruction

After an unseen first pass during training, spend no more than thirty seconds reconstructing five things without looking. This drill tests whether the passage exists in memory as relationships rather than fragments.

1. Object: What kind of text was it, and what was the broad issue or situation?
2. Viewpoint: Who controlled the main perspective, and which other voices mattered?
3. Movement: What were the two to five major stages?
4. Turn: Where did stance, mood, knowledge, scale or relationship change most?
5. End state: What position, condition, decision or unresolved tension remained at the end?

Then reopen the passage and check. Missing a minor example is not automatically a weakness. Missing the main concession, final position or voice separation is. The reconstruction therefore evaluates structural memory rather than total recall.

Repeat the drill across different genres. Over several weeks, the learner should need fewer written supports to achieve the same accuracy. That reduction indicates that passage architecture is becoming internalised.

Technical Appendix G: First-read quality scorecard

A teacher or self-studying learner can score a first read on five dimensions from 0 to 2. The purpose is diagnostic consistency, not another high-stakes grade.

• Structure: 0 = fragments only; 1 = broad sequence; 2 = functional movement and turn identified.
• Viewpoint: 0 = voices blended; 1 = main speaker clear; 2 = main and reported positions accurately separated.
• Scope: 0 = key qualifications lost; 1 = some retained; 2 = conditions, certainty and quantity preserved where important.
• Uncertainty control: 0 = either guessing or stopping excessively; 1 = mixed; 2 = unknowns triaged and structural breaks repaired locally.
• Navigation: 0 = cannot predict where to return; 1 = rough locations; 2 = major evidence regions can be located from the map.

A total score matters less than the profile. A learner with strong structure and poor scope needs different work from one with accurate scope but no navigation. Repeated scorecards across unfamiliar passages show whether the system is becoming robust rather than merely succeeding on one text.

Technical Appendix H: The handoff from reading the passage to answering the paper

The passage map is not an answer bank. It is a routing layer. Once questions begin, the learner should stop trying to make the map increasingly detailed and instead use it to choose where close reading is worth spending time.

A question about a local phrase sends the learner from wide view to close view. A question about development sends the learner to two or more mapped stages. A question about writer attitude sends the learner to stance changes and qualification. A question about comparison activates separate source maps and then a shared axis. The map narrows search; the exact question determines the final evidence corridor.

These guides address successive steps in comprehension. SETC-SEC-COMP-020 teaches how to make the unseen text navigable. SETC-SEC-COMP-010 teaches how a specific question becomes a mark. The broader English Reading Performance guide remains the foundation for evidence, inference and scope. The next guide explains how to manage time across these operations.

MAP WIDELY. RETURN NARROWLY. ANSWER EXACTLY.