OLSAT Level F common mistakes
OLSAT Level F · Understand the traps and how to avoid them.
On antonym items, the candidate picks the word that is merely unrelated to the prompt word rather than the word that is opposite in meaning (e.g., for 'scarce' choosing 'hidden' instead of 'abundant').
Under time pressure the mind latches onto any word that 'feels different' from the prompt; unrelatedness and opposition both produce a sense of contrast, and the candidate never explicitly tests the pair for a true opposite relationship.
What to do instead
Force a two-step check for every antonym option: (1) state the prompt word's core meaning in one phrase, (2) plug the option into the frame 'X means the opposite of Y' and say the sentence aloud mentally. If the frame produces a sentence that is false or only loosely true, eliminate. Practise with sets of five options where exactly one is a true opposite and two are merely unrelated, until the unrelated trap is recognised instantly.
On verbal analogy items, the candidate matches on surface topic (both words are about weather, both are animals) instead of matching the underlying relationship, so 'drizzle : downpour' gets paired with 'breeze : wind' rather than 'sprinkle : deluge'.
The brain categorises by theme faster than it analyses relationships; thematic overlap creates a false feeling of fit, and the candidate stops checking once an option 'sounds related'.
What to do instead
Before looking at options, write the bridge sentence linking the stem pair in a specific form ('a ___ is a weak version of a ___'). Then test each option by substituting it into that exact sentence and rejecting any option that requires changing the sentence's structure. If two options fit, the bridge was too vague - tighten it (add order, degree, or function) and retest.
On sentence-completion and classification items, the candidate selects an answer that fits the sentence grammatically but ignores a logical clue such as 'although', 'despite', or 'therefore', producing a sentence that contradicts itself.
Grammatical plausibility is processed automatically and creates a strong 'this sounds right' signal; contrast and causal connectives are small words that are easy to skim past, especially when the blank is far from the clue.
What to do instead
Adopt a routine: circle or mentally flag every connective and negative word ('not', 'except', 'although', 'unless') before reading the options, and decide whether the blank must agree or contrast with the rest of the sentence. After choosing, reread the full completed sentence and check that the connective's logic holds. For classification items, state the shared rule in words before examining the choices.
On verbal reasoning items with multi-step constraints, the candidate builds a single arrangement, finds it conflicts with one clue, and abandons the whole approach instead of adjusting the one violated position.
Working memory is overloaded by holding several clues at once; the first conflict is interpreted as 'my method is wrong' rather than 'one placement is wrong', which triggers restarting from scratch and burns time.
What to do instead
Use a compact notation (initials plus arrows, or a simple grid) and apply clues one at a time, checking after each addition. When a conflict appears, locate which clue is violated and move only the element involved. Practise short constraint sets (three to four items) until incremental repair becomes the default habit rather than full restart.
On pictorial reasoning items involving folding, punching, or paper transformations, the candidate answers from the folded state or forgets that holes are mirrored when the paper is unfolded, producing an answer with the correct number of holes but the wrong positions.
The folded figure is the last thing seen, so it dominates the mental image; mirror symmetry across the fold line is an extra transformation that is easy to skip, and counting holes feels like sufficient verification.
What to do instead
Unfold step by step in reverse order, drawing the mirrored copy across each fold line explicitly rather than imagining it. Verify by checking both hole count and symmetry: each hole must have a mirror partner across every fold line it crosses. Do three or four practice folds with a real sheet of paper to calibrate the mental model, then switch to drawing only.
On series and matrix items, the candidate identifies one rule (e.g., the shape rotates 90 degrees) and stops, missing a second simultaneous rule such as shading alternating or the inner element changing size.
Finding any pattern produces closure and reduces the discomfort of uncertainty; the candidate then selects the first option consistent with the single rule found, without scanning for a competing rule that would eliminate it.
What to do instead
Run a fixed checklist on every series or matrix: position, orientation, size, shading/fill, number of elements, and inner-versus-outer. State each attribute's behaviour explicitly, even 'unchanged'. Only then compare options, and eliminate any option that violates a confirmed rule before choosing among survivors.
On figural analogy and classification items, the candidate applies the transformation in the wrong direction or to the wrong element (rotating the whole figure when only the inner shape rotated), so the answer is a plausible but incorrect variant.
Transformations are perceived holistically; the candidate registers 'something changed' without isolating which component changed and how, and direction (A to B versus B to A) is easily reversed under time pressure.
What to do instead
Describe the stem change in a precise sentence naming the component and the operation ('the inner triangle rotates 90 degrees clockwise while the outer square stays fixed'). Apply that sentence to the third figure, then check the chosen option against the sentence in reverse to confirm it maps back correctly.
The candidate spends disproportionate time on one hard item early in a section and leaves later, easier items unanswered when time expires.
Persistence is rewarded in schoolwork, so abandoning a problem feels like failure; sunk cost keeps the candidate reworking the same item, and there is no pre-set time limit per question.
What to do instead
Set a personal cap per item (roughly one minute, or two for multi-clue puzzles) and rehearse a physical exit routine: mark the item, guess a placeholder answer, and move on. Since most of these tests have no penalty for wrong answers, never leave a blank - reserve the final minutes for returning to marked items.
The candidate answers the question they expected rather than the one asked - for example, choosing the synonym when the item says 'opposite', or picking the figure that comes next when asked which one does not belong.
Item types are interleaved and share surface features; the expectation built from the previous question carries over, and the single instruction word is read quickly and not re-checked.
What to do instead
Before answering, underline or mentally repeat the instruction keyword (opposite, same, next, does not belong, except). After selecting an answer, ask one final question: 'Does this answer the instruction that was actually written?' This two-second verification catches most misread items.
On vocabulary-dependent items, the candidate eliminates an option simply because the word is unfamiliar, treating 'I don't know this word' as evidence that it is wrong.
Unfamiliar words feel risky, and the candidate confuses low confidence with low plausibility; the result is a bias toward the most common or shortest option rather than the correct one.
What to do instead
Use word-part analysis before eliminating: prefix, root, and suffix (bene-, mal-, -cede, -ject, -ous, -ify) often reveal enough meaning to judge the option. If a word remains opaque, keep it as a live candidate and decide by eliminating options that are clearly wrong, rather than by familiarity. Build a running list of missed roots and review it weekly.