Irish Logical Thinking common mistakes
Irish Logical Thinking · Understand the traps and how to avoid them.
Reading a verbal passage's question stem first and then hunting for a matching keyword in the text, choosing the option whose wording echoes the passage rather than the one that answers the question.
Keyword matching feels efficient under time pressure and the exam's verbal items often reuse passage vocabulary in distractors, so surface overlap masquerades as correctness.
What to do instead
Read the stem, then restate the required relationship in your own words before looking at options (e.g. 'the author claims X causes Y, not merely that X and Y co-occur'). Eliminate any option that only repeats passage wording without asserting that relationship, and justify the survivor in one sentence.
Treating a 'which of the following is NOT supported' or 'EXCEPT' item as a normal positive item and selecting the first true-sounding statement.
Polarity words are short and easy to skim past, and the brain defaults to confirming rather than falsifying under time pressure.
What to do instead
Circle or mentally capitalise the negative or exception word before reading options, then test each option against the passage and select the one that fails the test. Reserve time to re-check polarity on every flagged item before moving on.
Accepting a conclusion that goes beyond the evidence, e.g. inferring a general rule from one example or assuming causation from a stated correlation.
Everyday reading habits fill gaps with plausible assumptions, and distractors are written to match those assumptions rather than the text.
What to do instead
After choosing an answer, label the evidence type it relies on: stated fact, single instance, correlation, or opinion. Reject any option requiring a stronger evidence type than the passage supplies, and practise tagging inference strength on a set of ten items.
Estimating percentages and ratios by feel, e.g. deciding that a 20% increase followed by a 20% decrease returns to the original value, or comparing 3/7 and 4/9 without computing.
Percentage change is asymmetric and intuition treats it as reversible; fractions with close values are hard to rank visually.
What to do instead
Use a fixed base of 100 for percentage chains (100 → 120 → 96) and cross-multiply to compare fractions. Write the intermediate number down rather than holding it in memory, and sanity-check direction of change before selecting.
Solving multi-step word problems with only the final answer in mind, so unit mismatches (minutes vs hours, cents vs euro) and the actual question asked are missed.
Working memory is consumed by the arithmetic, and the last number computed feels like the answer even when the question asked for a rate or a remainder.
What to do instead
Before calculating, write the target quantity with its unit (e.g. 'answer in km/h'). Convert all given values to consistent units at the start, and after computing, check the result's unit and rough magnitude against the problem's context.
Spending equal time on every numerical item, getting stuck on one data-heavy question and losing three or four easier marks later in the section.
Sunk-cost pressure makes abandoning a nearly-solved problem feel like waste, and candidates underestimate how long table-reading items take.
What to do instead
Set a personal time cap per item (roughly one minute for short items, two for table or chart items). If the cap passes, mark and move on; return only after all items are attempted. Rehearse this with timed sets so the abandon decision is practised, not improvised.
Completing a series or matrix by pattern-matching to a familiar rule (alternating, +2, rotation) without verifying that the rule accounts for every given element.
The first plausible rule produces cognitive closure, and abstract items are designed so a partial rule fits the first two or three elements.
What to do instead
State the candidate rule explicitly and test it against the last given element before applying it. If it fails anywhere, discard it entirely rather than patching it. For matrices, check rows and columns separately and confirm the rule holds in both directions.
Losing track of orientation and position when a figure is rotated or reflected, especially confusing a 90-degree clockwise rotation with a reflection across a diagonal.
Mental rotation is error-prone for asymmetric shapes, and diagonal reflection and quarter-turn rotation produce superficially similar results.
What to do instead
Track one distinctive feature (a marked corner or a notch) rather than the whole shape, and note its before/after position. For reflection items, check chirality: a reflection reverses handedness, a rotation never does. Verify with a second feature before answering.
Drawing a conclusion from syllogistic premises by matching the words rather than testing validity, so 'All A are B; some B are C' is taken to imply 'some A are C'.
Natural language encourages assuming that shared terms link the categories, and the invalid conclusion sounds fluent.
What to do instead
Use a quick diagram or counterexample method: place the categories as circles and check whether the conclusion must hold in every arrangement consistent with the premises. If one arrangement makes it false, the conclusion does not follow. Apply the same test to 'some' and 'no' statements, which behave differently from 'all'.
Treating conditional statements as biconditional, inferring 'if P then Q' gives 'if Q then P', or denying the antecedent to conclude the negation of the consequent.
Conversational implicature makes the reverse direction feel implied, and the two fallacies are the most common invalid forms in deduction items.
What to do instead
Rewrite each conditional in the form 'P → Q' and note that only two inferences are valid: affirm P to conclude Q, or deny Q to conclude not-P. Label each option as one of the four forms before judging it, and drill the contrapositive until it is automatic.
Solving scheduling, seating, or grouping puzzles by holding constraints in memory and placing items one at a time, then discovering a violated constraint after several placements.
Constraint problems overload working memory, and early placements made without checking all affected rules force costly backtracking.
What to do instead
Externalise the setup: draw the grid or timeline and list every constraint as a numbered rule. Place the most restrictive item first, tick off which rules each placement satisfies, and check all rules after every placement rather than at the end.