Reading Trap Taxonomy

SAT wrong answers aren't random; they're a small set of repeating, engineered trap types — once you can name the trap, you eliminate it without agonising. This vault catalogues the distractor patterns that run through every R&W reading question. Pair it with the FORGE type-systems (method) and the Error Log (to see which traps catch you).

Sources: Meltzer; UWorld; PrepMaven; ArborBridge. Examples MERIDIAN-original.

---

#THE TRAP CATALOGUE

T1 · Too-extreme. A choice stronger/more absolute than the text supports (always, never, proves, all, impossible). Tell: absolute language on a passage that hedges. Appears: inference, central idea, evidence.

T2 · One step too far. True-adjacent but adds a claim the text doesn't make (a prediction, a generalisation, a cause). Tell: you have to assume one more thing to make it true. Appears: inference (its signature), evidence.

T3 · Correlation → causation. Text shows two things co-occur; the trap says one causes the other. Tell: a causal verb (causes, leads to, because of) where the text only links. Appears: inference, quantitative evidence, data.

T4 · Outside knowledge. True in the real world, not stated/implied here. Tell: you "know" it but can't point to a line. Appears: inference, central idea.

T5 · On-topic, wrong claim. Relevant to the subject, irrelevant to the specific claim asked about. Tell: it's about the right topic but answers a different question. Appears: command of evidence (textual), purpose.

T6 · True but irrelevant. Accurately stated (often straight from a graph) but doesn't do the job the question needs. Tell: "yes, that's true… but it doesn't answer this." Appears: quantitative evidence, detail.

T7 · Reversed / direction flip. Swaps the direction of a relationship or which author/variable holds which value. Tell: right elements, wrong order. Appears: cross-text, data, transitions.

T8 · Half-right. First clause correct, second clause wrong (or vice versa). Tell: you stopped reading the choice at the part that matched. Appears: everywhere — read the whole choice.

T9 · Right info, wrong purpose. Fluent, factually accurate, built from the notes — but serves a different rhetorical goal than the one stated. Tell: great sentence, doesn't do the stated job. Appears: rhetorical synthesis (its signature).

T10 · Primary-meaning bait. The common meaning of a word used here in a secondary sense; or a synonym-in-isolation that breaks the sentence's logic. Tell: the "obvious" word; or two near-synonym options (both then wrong). Appears: words in context.

T11 · One-passage answer. True of Passage 1 or 2 alone, not of the relationship asked about. Tell: it describes one author, not the link. Appears: cross-text connections.

T12 · Invented agreement / overstated conflict. Claims the texts agree where they don't, or that they clash where one only qualifies the other. Tell: the relationship is stronger than the texts support. Appears: cross-text.

T13 · Distorted paraphrase. Restates a detail but changes one word that changes the meaning. Tell: sounds like the text but a key term shifted. Appears: detail, central idea.

T14 · Scope error. Describes the whole passage when asked about one sentence (or vice versa). Tell: right content, wrong zoom level. Appears: purpose/structure.

---

#THE UNIVERSAL ELIMINATION HABIT

For every choice you're tempted by, ask: "which trap could this be?" If you can name a trap (T1–T14), eliminate it. The correct answer is the one with no nameable defect — usually the most cautious, most text-anchored, least flashy option. Why (mechanism): naming converts a vague "this feels off" into a decisive elimination, and stops you choosing the interesting answer over the supported one. [This is the difference between 90% and 100% accuracy on reading.]

#DIAGNOSE (via your Error Log)

Tag each reading miss with its trap number. After 2–3 tests, your top 2–3 trap numbers are your training targets — drill those question types with the trap in mind.

#DRILL (self-marked · MERIDIAN-original)

Trap-naming (target 8/8, ≤4 min): for 8 wrong answers (from any practice set), write the trap number and its tell. PASS: correct trap named on ≥7/8. Re-run weekly; watch your recurring trap numbers shrink.

#REFERENCE CARD — READING TRAPS

T1 too-extreme · T2 one-step-too-far · T3 correlation→causation · T4 outside knowledge
T5 on-topic-wrong-claim · T6 true-but-irrelevant · T7 reversed · T8 half-right
T9 right-info-wrong-purpose · T10 primary-meaning bait · T11 one-passage · T12 invented agreement
T13 distorted paraphrase · T14 scope error
Ask of every tempting choice: "which trap is this?" Name it → eliminate it.
Correct = the option with NO nameable defect (cautious, text-anchored).

---

MERIDIAN™ · © 2026 · For Neo's personal study use. Not affiliated with or endorsed by the College Board. SAT® is a trademark of the College Board.

MERIDIAN · CODEX · READING TRAP TAXONOMY | v1.0