Module 2 · UPSC CSAT (Paper 2)

Logical Reasoning

Updated 18 Nov 2026 · 10,159 words · illustrated
Updated 14 Nov 2026
Analogies, syllogisms, puzzles, seating arrangement, blood relations, directions.
Analogies · Syllogisms · Puzzles · Blood Relations · Directions
🎯 Take Mock Test
PremiseIdentificationLogical StructureInference RulesConclusionValidationfeedsfeedsappliesGS1 Logical Reasoning – Syllogism FlowArt. 51A(h): Develop scientific tempervlymbooq
Syllogism Flow for GS1 Geography Logical Reasoning Exam — Shows the step‑by‑step syllogism process used in GS1 logical reasoning questions. · Open the full diagram →
ObservationPattern RecognitionHypothesisTestingConclusionleads toleads toleads toleads toGS1 Geography: Logical Reasoning ChainUse: Observe → Pattern → Hypothesize → Test → Concludevlymbooq
Logical Reasoning Steps Applied to Geography Exam Questions — Shows the sequential reasoning chain from observation to conclusion used in GS1 geography questions. · Open the full diagram →

Learning Objectives

  • Understand the fundamental concepts of Logical Reasoning
  • Apply key formulas and techniques to solve problems
  • Practice with exam-level questions to build speed and accuracy

1. What is Logical Reasoning?

Logical reasoning in UPSC CSAT tests your ability to analyse patterns, draw conclusions from given information, and solve problems using structured thinking. Topics include analogies, syllogisms, puzzles, blood relations, directions, series, and data sufficiency. CSAT typically has 8-12 questions from this area.

Key Concepts

1. Analogies: Identify the relationship between a given pair and apply it to find the missing term. Types: word (synonym, antonym, cause-effect, part-whole), letter (positional value, shifting patterns), number (arithmetic operations, special sequences).

2. Syllogisms: Use Venn diagrams to test conclusions. Statements like 'All A are B', 'No A is B', 'Some A are B', 'Some A are not B'. Draw all possible Venn diagrams — a conclusion follows only if true in ALL possible diagrams.

3. Blood Relations: Family tree problems. Common relationships: father, mother, brother, sister, husband, wife, son, daughter, uncle, aunt, cousin, nephew, niece. Draw arrows showing parent-child relationships and label each person clearly.

4. Direction & Distance: Track movements using cardinal directions (N, S, E, W). Use Pythagoras theorem for displacement: d² = (total N-S)² + (total E-W)². Right turn = 90° clockwise. Left turn = 90° anticlockwise.

5. Series: Number series — identify pattern (AP, GP, Fibonacci, alternating operations, digit patterns). Alphabet series — positional values with addition/subtraction patterns.

6. Seating Arrangement: Linear, circular, or rectangular arrangements. Use a diagram and fill positions step by step based on given conditions. Left/right depends on whether people face centre (circular) or a fixed direction (linear).

Solved Examples

Example 1 (Syllogism)
Statements: All fruits are sweet. Some sweets are healthy. No healthy is sour.
Conclusions: I. Some fruits are healthy. II. No sweet is sour.

a) Only I follows b) Only II follows c) Both follow d) Neither follows
Solution: d) Neither follows. Conclusion I: All fruits are sweet (inside circle of sweets). Some sweets are healthy — this means the healthy circle overlaps with sweets. But fruits could be entirely outside the healthy area. So I does not follow necessarily. Conclusion II: No sweet is sour. We know that some sweets are healthy, and no healthy is sour. But sweets that are not healthy could be sour. So II does not follow necessarily.
Example 2 (Blood Relation)
A is the sister of B. B is the mother of C. C is the brother of D. E is the father of D. How is A related to D?

a) Aunt b) Mother c) Grandmother d) Sister
Solution: a) Aunt. A is sister of B. B is mother of C and D (since C is brother of D, they share the same mother B). E is father of D — so E is B's husband (or partner). A, being B's sister, is the aunt of D. Family tree: E married to B, B's sister is A. B and E have children C and D. So A is D's aunt.
Example 3 (Direction & Distance)
Raj walks 15 m North. He turns right and walks 20 m. He turns right and walks 15 m. He turns left and walks 10 m. How far and in which direction is he from the starting point?

a) 10 m East b) 20 m East c) 10 m West d) 30 m East
Solution: d) 30 m East. Start at O(0,0). Walk 15 m North → (0,15). Turn right (East), walk 20 m → (20,15). Turn right (South), walk 15 m → (20,0). Turn left (East), walk 10 m → (30,0). Net displacement: 30 m East.

2. Shortcut Techniques

  • For syllogisms, always draw Venn diagrams — do not rely on intuition alone. Draw all possibilities and check each conclusion against every diagram.
  • In blood relations, use a standard family tree notation: male = square, female = circle, marriage = horizontal line, siblings = vertical branch.
  • For direction problems, draw a rough sketch instead of mental calculation — it prevents mistakes and saves time in complex multi-step problems.
  • For seating arrangement, start with the most specific condition first (direct statements like 'A sits at the left end') before handling relative conditions.
  • In number series, check differences, ratios, squares, cubes, and alternating patterns in that order — most series follow one of these patterns.
Pro Tip
CSAT logical reasoning is at class 10 level — it tests speed and accuracy, not complexity. If a puzzle seems too complex, you are likely overcomplicating it. Look for simple patterns and direct relationships. Practice 5-10 questions daily to build speed.

3. Comprehensive Study Guide

Master logical reasoning with regular practice and concept clarity.

4. Practice Strategy

Phase 1: Learn concepts. Phase 2: Solve 30-40 basic questions. Phase 3: Attempt mixed-level questions. Phase 4: Take timed mocks. Phase 5: Analyze and improve.

Pro Tip
Consistency beats intensity. Study 1-2 hours daily rather than 6-8 hours on weekends.

5. Syllogistic Logic: The Three-Term Constraint and Venn Diagram Rigor

The fundamental mechanism of syllogism in competitive examinations relies on adherence to the "Three-Term Rule". A valid syllogism must contain exactly three distinct terms: the Major Term (predicate of the conclusion), the Minor Term (subject of the conclusion), and the Middle Term (which appears in both premises but not in the conclusion). The middle term acts as the bridge; if it is not distributed at least once, the argument commits the fallacy of the undistributed middle. For instance, consider the statements: "Some cats are black" and "Some black things are dogs." A hasty conclusion that "Some cats are dogs" is invalid because 'black' is the middle term, but it refers to a subset in both premises (undistributed), failing to connect the two groups effectively.

Examiners frequently test candidates on the distinction between "All", "No", and "Some". In formal logic, "Some" implies "at least one", not necessarily "more than half" or "most". Furthermore, the position of terms matters. If a premise states "All A are B", it does not imply "All B are A" (Fallacy of Affirming the Consequent). This is often tested using Venn diagrams where the circle for A is drawn entirely inside the circle for B. However, if the statement is "Some A are B", the circles must intersect, and one cannot assume the other parts of the circles are empty or full without further information.

A critical exception often overlooked involves negative premises. If both premises in a syllogism are negative (e.g., "No A is B" and "No C is B"), no valid conclusion can be drawn between A and C. This is because there is no positive link established to bridge the gap. Similarly, if one premise is negative, the conclusion must necessarily be negative. If you draw a Venn diagram where two circles are disjoint (separate), you cannot infer any relationship between their contents unless an external condition connects them. The visual representation in a Venn diagram must cover all possible scenarios: for "Some A are B", you must draw the intersection, but you must also acknowledge that parts of A and B outside the intersection exist.

The concept of "Possibility" is another high-yield area. Statements like "It is possible that some A are B" do not require a definitive conclusion to follow; they simply ask if there exists *any* scenario consistent with the premises where the statement holds true. In such cases, you must construct a Venn diagram that satisfies the conditions and check if the proposed possibility fits within that constructed reality. If the premises allow for an overlap even if it is not explicitly stated in the text, the possibility conclusion follows. This differs from definite conclusions which must hold true in *every* possible arrangement.

Examiner's Trap: The Hidden Assumption
Question: Statements: 1. All roses are flowers. 2. Some flowers are red. Conclusion I: Some roses are red. Conclusion II: No rose is red. Options: a) Only I follows b) Only II follows c) Either I or II follows d) Neither follows.
Solution: Option (c). Analysis: From Statement 1, the circle of Roses is inside Flowers. From Statement 2, the Red circle overlaps with Flowers. Since Roses and Red both overlap with Flowers, they can either overlap each other or remain separate. Therefore, it is possible that some roses are red, AND it is also possible that no rose is red (if the red part of flowers does not touch roses). In logic questions asking for "Either/Or" cases, if two conclusions cover all possibilities and cannot both be true simultaneously, the answer is 'Either I or II'. Here, 'Some roses are red' and 'No rose is red' are contradictory pairs regarding the specific subset, making 'Either/Or' the correct logical inference.
Exam-Day Strategy: The Venn Diagram Shortcut
Never solve a syllogism purely by intuition. Always draw three circles (or two if one term is absent) on scratch paper. Use a standardized notation: draw the intersection for 'Some', draw one inside another for 'All', and keep them separate for 'No'. If a conclusion requires an overlap that your diagram shows as non-existent, it is false. Spend no more than 45 seconds per syllogism; if you cannot visualize the diagram quickly, mark it as 'Neither' and move on to save time for complex puzzles.

6. Blood Relations: Decoding Complex Family Structures and Gender Ambiguities

Solving blood relation problems requires a systematic approach to mapping kinship terms. The core difficulty lies in handling gender ambiguity and the distinction between biological and adoptive relationships, though UPSC typically focuses on biological lineage unless specified. A robust method is to start from the most definitive relationship given (e.g., "A is the father of B") and build the family tree outward or inward depending on the question's target. When a problem states "Pointing to a photograph, Manish said, 'She is the daughter of my mother\'s only son'," one must parse the pronouns carefully. If Manish is male (implied by context or name), his mother's only son is Manish himself. Thus, the person in the photo is the daughter of Manish.

A frequent source of error involves the terms "Sister", "Brother", and their cousins versus nephews/nieces. For example, if X is the brother of Y, and Z is the sister of X, then Z is also the sister of Y (half-siblings or full siblings depending on parents). However, in competitive exams, unless specified as half-siblings, siblings share both parents. The term "Cousin" implies a relationship through grandparents, not parents. A common trick involves the phrase "son of my father\'s son". If I say this about myself, it refers to me (since my father's son is me). If I have a brother, and I refer to "my father's son", it could be me or my brother, creating ambiguity that must be resolved by checking if the relationship allows for multiple possibilities. Usually, questions are framed to have a single definitive answer.

Another layer of complexity arises in step-parent relationships versus adoptive ones. While standard logic puzzles assume biological relations, real-world scenarios distinguish between "step-mother" and "mother". In exam contexts, unless the word "step" or "adopted" is explicitly used, assume biological connection. However, questions may ask: "How is A related to B?" where A is the wife of B's father. This makes A the step-mother, but if the question asks for the legal relationship in a strict sense without assuming marriage validity, it might be treated differently. Always look for keywords like "wife", "husband", "daughter-in-law" which denote marital relationships, distinct from blood ties. The hierarchy is: Parents -> Siblings -> Children (Nephews/Nieces) -> Cousins.

Consider a scenario involving multiple generations: "The mother of my only brother is the only daughter of your father." If you are male, your father's only daughter is your sister. Therefore, the mother of your brother (who is also your sister) is your sister? No, that implies incest or error. Let's re-parse: "My only brother" implies I am female (since if I were male, I would have a brother and myself are sons).

7. Analogical Reasoning: Structural Mapping and Multi-Level Patterns

Analogical reasoning tests the ability to perceive a relational structure between two pairs of terms and replicate it with a new pair. The core mechanism is structural mapping: identify the abstract relation (cause‑effect, part‑whole, degree, function, sequence, or opposition) binding the first pair, then select the answer choice exhibiting an isomorphic relation. In competitive examinations, analogies often extend beyond a single semantic layer; they embed multi‑level patterns where a primary relation is modified by a secondary qualifier such as intensity, directionality, or temporal order.

Consider the analogy Pen : Write :: Knife : Cut. The primary relation is instrument‑action. A multi‑level variant might be Pen : Write :: Scalpel : Incise, where a secondary layer adds professional context (medical vs. general). Examiners exploit this by offering distractors that satisfy the primary relation but violate the secondary qualifier (e.g., Knife : Slice – correct primary relation, but Slice lacks the precision connotation of Incise). Recognising the qualifier requires parsing lexical nuance: register (formal vs. colloquial), domain (scientific, domestic, legal), and valence (positive, negative, neutral).

A systematic approach uses a relation‑template table. List the first pair, articulate the relation in a single clause (e.g., “X is the tool used to perform Y”), then test each option by inserting its terms into the same clause. Options yielding a grammatically and semantically coherent sentence survive; the rest are eliminated. When two options survive, compare secondary attributes: morphological complexity, collocation frequency, and domain specificity. The option mirroring the original pair on the most attributes is correct.

Common analogy families include:

  • Synonym/antonym gradients – Warm : Hot :: Cool : Cold (degree escalation).
  • Part‑to‑whole – Petal : Flower :: Leaf : Tree (component‑system).
  • Cause‑effect – Rain : Flood :: Spark : Fire (trigger‑outcome).
  • Function‑user – Stethoscope : Doctor :: Chalk : Teacher (instrument‑profession).
  • Sequence‑stage – Larva : Pupa :: Seed : Sapling (developmental phase).

Advanced items may combine families, e.g., Architect : Blueprint :: Composer : Score (creator‑plan) where both terms belong to the creative‑design domain. The examinee must map both the structural relation and the domain tag. Practice with timed drills requiring explicit relation statements reduces reliance on intuition and exposes hidden qualifiers.

8. Advanced Syllogistic Forms: Existential Import and Distributed Terms

Standard syllogisms often adopt the Boolean interpretation where universal statements (All A are B, No A are B) carry no existential import; they do not assert the existence of any A. However, some examination questions operate under the Aristotelian interpretation, which treats universal affirmatives as implying at least one member of the subject class. This distinction changes the validity of conclusions such as Some A are B derived from All A are B. Under Boolean logic the inference is invalid; under Aristotelian logic it is valid. Candidates must detect the convention from wording: phrases like “Assume the category exists” or “There is at least one” signal Aristotelian import.

Distribution of terms governs permissible inferences. A term is distributed when the statement makes a claim about every member of that class. In All A are B, subject A is distributed, predicate B is not. In No A are B, both are distributed. In Some A are B, neither is distributed. In Some A are not B, predicate B is distributed. The distribution rule for a valid categorical syllogism: the middle term must be distributed at least once; any term distributed in the conclusion must be distributed in the premises. Violations produce the fallacies of undistributed middle, illicit major, and illicit minor.

Conversion, obversion, and contraposition are immediate inferences preserving truth value under specific conditions. Conversion (switching subject and predicate) is valid for E (No A are B → No B are A) and I (Some A are B → Some B are A) but not for A or O. Obversion (changing quality and replacing predicate with its complement) is valid for all four forms. Contraposition (replacing both terms with complements and switching) is valid for A and O but not for E and I. Mastery of these transformations allows rapid reduction of complex premises to a standard three‑term structure before Venn diagramming.

When premises exceed three terms, the chain syllogism technique applies: link premises by shared terms to form a chain (A→B, B→C, C→D) and derive the transitive conclusion (A→D). Each link must satisfy distribution rules. If any link fails, the chain collapses. Examiners often embed a hidden four‑term fallacy by using synonyms that appear distinct (e.g., “physicians” and “doctors”) – treat them as identical only when the question explicitly equates them.

Worked Syllogism with Existential Import
Statements: 1. All engineers are problem‑solvers. 2. Some problem‑solvers are creative. Conclusions: I. Some engineers are creative. II. No engineer is creative. Options: a) Only I follows b) Only II follows c) Either I or II follows d) Neither follows
Step 1: Identify the interpretation. The statements contain no explicit existential qualifier, so adopt the Boolean interpretation (standard for most competitive exams). Step 2: Diagram. Draw three circles: Engineers (E), Problem‑solvers (P), Creative (C). Statement 1: E entirely inside P. Statement 2: Overlap between P and C (some). Step 3: Test Conclusion I – “Some E are C”. The diagram shows E inside P, and P overlaps C, but the overlap may lie completely outside E. Hence “Some E are C” is not forced; it is possible but not necessary. Conclusion I does not follow. Step 4: Test Conclusion II – “No E are C”. The diagram permits E to intersect C (if the P‑C overlap includes part of E). Therefore “No E are C” is not forced either. Step 5: Since neither conclusion is logically necessary, the correct answer is (d) Neither follows. Note: Under Aristotelian import, “All E are P” would imply existence of at least one E, but that still would not guarantee overlap with C. The result remains (d).
Exam‑Day Syllogism Protocol
1. Read the question stem for existential cues. 2. Write the three term symbols (S, M, P) on scratch paper. 3. Mark distribution for each premise using the mnemonic “A distributes Subject, E distributes Both, I distributes None, O distributes Predicate”. 4. Apply the two distribution rules; if violated, answer “Neither” immediately. 5. If rules pass, draw a minimal Venn diagram (three circles) and shade/place X marks. 6. Evaluate each conclusion against the diagram. 7. Allocate maximum 45 seconds; if diagram is ambiguous, choose “Neither” and flag for review.

9. Puzzle Architecture: Constraint Propagation and Grid Deduction

Logical puzzles in competitive tests are essentially constraint satisfaction problems (CSPs). Each entity (person, object, event) possesses a set of attributes (colour, profession, floor, day) and a set of binary constraints (relative ordering, exclusivity, conditional links). The solving process is constraint propagation: repeatedly apply deterministic deductions (e.g., “A is not on floor 3”) to prune the domain of each variable until a unique assignment emerges or a branching point forces hypothesis testing.

A robust representation is the grid matrix (logic grid). Rows correspond to entities, columns to attribute categories. Cells hold three states: confirmed (✓), eliminated (✗), or undetermined (blank). The algorithm proceeds in passes:

  1. Direct clues – Fill confirmed/eliminated cells from explicit statements (“The doctor lives on floor 2”).
  2. Negative clues – Apply “not” statements (“The engineer is not on floor 1”).
  3. Relative clues – Translate ordering (“X lives immediately above Y”) into pairwise constraints restricting possible floor pairs.
  4. Conditional clues – Encode “If P then Q” as implication; when P is confirmed, enforce Q; when Q is eliminated, eliminate P (contrapositive).
  5. Global constraints – Enforce uniqueness: each attribute value used exactly once per category (one person per floor, one colour per house).

After each pass, scan for singleton domains – a row or column with only one blank cell; that cell must be confirmed. This triggers a new pass. The process iterates until the grid is complete or a branching variable (a cell with two equally possible states) appears. At that point, adopt a depth‑first hypothesis: assume one state, propagate, and check for contradiction. If contradiction arises, revert and adopt the alternative. Because puzzle sizes are small (typically 4‑6 entities), a single branching level usually suffices.

Advanced puzzles introduce multi‑dimensional grids (e.g., person × floor × pet × drink). The same propagation logic applies, but the matrix expands to a 3‑D tensor. A practical shortcut is to maintain separate 2‑D grids for each pair of dimensions and cross‑reference them. For instance, a “person‑floor” grid and a “person‑pet” grid share the person dimension; a confirmation in one grid instantly updates the other.

Common traps: over‑reliance on memory – always record every deduction on the grid; mental tracking leads to missed eliminations. Misreading “immediately” vs. “somewhere above” – “Immediately above” fixes distance to one unit; “above” only fixes order. Mark the distinction with a symbol (↑1 vs. ↑).

10. Direction Sense: Vector Decomposition and Relative Motion

Direction sense questions model movement on a planar grid using cardinal (N, S, E, W) and inter‑cardinal (NE, NW, SE, SW) bearings. The underlying mathematics is vector addition in a Cartesian coordinate system where North = +y, East = +x. A movement of d metres at bearing θ (measured clockwise from North) decomposes into Δx = d·sinθ, Δy = d·cosθ. For cardinal directions, sin and cos reduce to 0, ±1, simplifying computation.

Standard problem types:

  • Net displacement – Sum all Δx and Δy components; final distance = √(Δx² + Δy²); final bearing = atan2(Δx, Δy) converted to compass notation.
  • Shadow problems – At sunrise (East), shadows point West; at sunset (West), shadows point East. At noon (sun overhead), shadow length is minimal. The direction of a person’s shadow reveals the person’s facing direction relative to the sun.
  • Relative motion – Two moving objects: compute relative velocity vector by subtracting one velocity vector from the other. The bearing of the relative vector indicates the direction of one as seen from the other.
  • Turn sequences – Left/right turns change heading by ±90° (or ±45° for half‑turns). Track heading as an angle modulo 360°; each forward step adds the current heading’s unit vector.

Efficient solving uses a coordinate table. Initialise (0,0). For each instruction, update coordinates. Example: “Walk 10 m North, turn right, walk 5 m, turn left, walk 10 m.” Steps: (0,10) → turn right (heading East) → (5,10) → turn left (heading North) → (5,20). Net displacement = √(5²+20²) ≈ 20.6 m, bearing ≈ arctan(5/20) ≈ 14° East of North (N14°E).

When distances are relative (“twice the previous distance”), assign a variable (e.g., first leg = a) and express subsequent legs in terms of a. The final coordinates become linear expressions in a; the distance simplifies to a·√(k) where k is a constant. The variable cancels if the question asks for direction only.

Examiner twists: non‑orthogonal grids (hexagonal or triangular lattices) appear rarely; treat them by defining basis vectors for the lattice directions. Mirror‑image traps – a description from the perspective of an observer facing South reverses left/right. Always re‑orient the narrative to a fixed North‑up frame before solving.

11. Blood Relations: Recursive Generation Mapping and Ambiguity Resolution

Complex blood‑relation puzzles require constructing a generation‑indexed family tree. Assign each individual a generation number (G0 = reference person, G‑1 = parents, G‑2 = grandparents, G+1 = children, etc.). Represent relationships as directed edges: parent→child (downward), spouse↔spouse (horizontal), sibling↔sibling (same generation, shared parents). This graph structure makes recursive queries tractable: “X’s mother’s brother’s son” translates to: start at X (G0), move to mother (G‑1), move to brother (G‑1, same parents), move to son (G0). The result is a male cousin of X (same generation, different parental line).

Gender ambiguity is the primary source of multiple valid trees. When a statement says “A is the sibling of B” without gender, both A and B could be male or female. The resolution strategy: enumerate all gender assignments consistent with explicit clues (e.g., “A is the father of C” fixes A as male). For each assignment, propagate gender through marriage edges (spouse of male is female, etc.). If a query asks “How is A related to B?” and the answer differs across valid assignments, the correct response is “Cannot be determined”. Examiners often include this option precisely for this scenario.

Recursive mapping also handles step‑relations and adoptive links when explicitly mentioned. Treat a step‑parent as a spouse of a biological parent with no blood edge to the child. Adoptive parent receives a “legal‑parent” edge distinct from “biological‑parent”. Queries asking “biological mother” vs. “mother” must be answered using the appropriate edge type. In the absence of “step” or “adopted”, assume biological.

A practical solving template:

  1. List all named persons.
  2. Create a table with columns: Person, Generation, Gender (M/F/?), Spouse, Father, Mother.
  3. Fill direct clues (father, mother, husband, wife, son, daughter).
  4. Apply sibling rule: persons sharing both father and mother are siblings.
  5. Iterate: whenever a parent is identified, assign generation = child generation – 1; whenever a child is identified, assign generation = parent generation + 1.
  6. Resolve gender: if a person is referred as “son” → male; “daughter” → female; “husband” → male; “wife” → female; “brother” → male; “sister” → female.
  7. After saturation, answer the target query by tracing the unique path in the graph. If multiple paths exist with different relationship labels, select “Cannot be determined”.

Advanced puzzles may embed circular references (e.g., “A’s grandson is B’s father”). Detect cycles by checking generation consistency; a cycle forces a generation contradiction unless the cycle spans an even number of generations (impossible in a strict biological tree). Such contradictions indicate a mis‑interpreted clue or a trick question where the answer is “Data insufficient”.

  • Assuming universal statements carry existential import without explicit cue leads to invalid conclusions.
  • Neglecting distribution rules in syllogisms produces illicit major/minor fallacies.
  • Treating “immediately above” and “above” as equivalent in seating or floor puzzles.
  • Ignoring the contrapositive of conditional clues during constraint propagation.
  • Using a single mental coordinate system for direction problems without re‑orienting to a fixed North reference.
  • Overlooking gender ambiguity in blood‑relation chains and forcing a single relationship label.
  • Failing to record every elimination on the logic grid, causing missed singleton deductions.

12. Multi‑Level Analogy Chains and Structural Isomorphism

Analogies in competitive examinations often cascade through three or four relational layers. A multi‑level analogy chain can be modelled as a sequence of binary relations R₁, R₂, …, Rₙ where each relation maps an element from domain Dᵢ to codomain Dᵢ₊₁. Structural isomorphism requires that the relational pattern (e.g., “part‑to‑whole”, “cause‑to‑effect”, “tool‑to‑function”) remains invariant across the chain. To solve such items, first encode each given pair as an ordered tuple (a, b) and annotate the relation type using a controlled vocabulary: part‑whole, member‑class, agent‑action, object‑attribute, sequence‑step. Next, verify that the same label applies to every adjacent pair. If a label changes, the chain is broken and the answer must be the option that restores uniformity. Questions often insert a “distractor pair” that shares surface similarity (same semantic field) but violates the structural label; the candidate who relies on lexical association rather than relational abstraction selects the distractor. Practice by constructing chains of length four from standard word‑lists (e.g., leaf : tree :: petal : flower :: key : keyboard :: byte : memory) and explicitly writing the relation label for each link. This habit forces the brain to treat analogy as a formal mapping problem rather than a vocabulary test.

When the chain contains a missing term, treat the problem as a constraint‑satisfaction task. Let the unknown be X. Write the two equations R₁(A, X) and R₂(X, B) where R₁ and R₂ are the known relation types on either side of X. Because the relation types are identical in a well‑formed chain, R₁ = R₂ = R. The solution set for X is the intersection of the pre‑image of B under R and the image of A under R. In most test items the intersection is a singleton, yielding a unique answer. If the intersection contains multiple candidates, the item is either flawed or expects “Cannot be determined”. Recognising this logical structure prevents guesswork and clarifies why some analogy questions have no unique answer.

13. Syllogistic Networks: Multiple Premises and Conditional Conclusions

Classical syllogisms involve exactly two premises and three terms. Examination papers, however, frequently present syllogistic networks: three or more premises that share terms in a graph‑like fashion, and a conclusion that may be conditional (e.g., “If all A are B, then some C are D”). The rigorous approach is to translate each categorical statement into its Boolean algebraic form using the four standard propositions: A (All S are P), E (No S are P), I (Some S are P), O (Some S are not P). Represent each term as a set variable; the universal affirmative A becomes S ⊆ P, the universal negative E becomes S ∩ P = ∅, the particular affirmative I becomes S ∩ P ≠ ∅, and the particular negative O becomes S ⊈ P. A network of premises yields a system of set inclusions, exclusions, and non‑emptiness constraints. Solving the network means determining whether the conclusion’s set expression follows logically from the system. This can be done by constructing a Venn diagram with one region per possible intersection of the involved terms (for n terms there are 2ⁿ regions) and shading or marking regions according to each premise. The conclusion is valid iff every model satisfying all premises also satisfies the conclusion. Conditional conclusions are handled by temporarily adding the antecedent as an extra premise, checking the consequent, and then retracting the antecedent. If the consequent holds under the augmented premise set, the conditional is valid. A common trap is to present a premise set that is individually consistent but jointly contradictory (e.g., “All A are B”, “All B are C”, “Some A are not C”). The contradiction forces the conclusion “No valid conclusion” or “Data insufficient”. Detecting inconsistency early — by checking whether any region is forced to be both empty and non‑empty — saves time and avoids the fallacy of drawing a conclusion from an impossible premise set.

Technique
Given the premises: (1) All engineers are analysts. (2) Some analysts are designers. (3) No designers are managers. (4) If all managers are engineers, then some engineers are not designers. Determine whether the conditional conclusion follows.
Translate: (1) E ⊆ A. (2) A ∩ D ≠ ∅. (3) D ∩ M = ∅. Conditional antecedent: M ⊆ E. Assume antecedent true. From (1) and antecedent, M ⊆ E ⊆ A, so M ⊆ A. From (3), D ∩ M = ∅, thus no manager is a designer. Since M ⊆ A, the set of managers is a subset of analysts that is disjoint from designers. Therefore there exists at least one analyst (any manager) that is not a designer, i.e., A ⊈ D, which is equivalent to “Some analysts are not designers”. Because every engineer is an analyst (1), some engineers are also not designers. Hence the consequent “some engineers are not designers” holds under the antecedent. The conditional conclusion is valid.
Exam‑Day Strategy
When a syllogistic network appears, sketch a quick Venn diagram with only the terms that actually occur in the premises and the conclusion. Shade universal negatives first, then mark particular affirmatives with an ‘×’ in any permissible region. If a conditional conclusion is asked, draw the antecedent as an extra shading/marking layer on a copy of the diagram; if the consequent’s required mark already exists, the conditional is true. This visual method is faster than algebraic manipulation and reduces careless errors.

14. Hybrid Puzzle Grids: Integrating Seating, Floor, and Attribute Constraints

Modern reasoning sections increasingly blend linear seating, multi‑floor building, and attribute assignment (colour, profession, vehicle) into a single puzzle. The underlying structure is a constraint‑satisfaction problem (CSP) with variables representing each person, each seat/floor position, and each attribute. Domains are finite sets (e.g., seats 1‑8, floors 1‑5, colours {Red, Blue, Green, Yellow}). Constraints fall into three families: binary relational (A sits immediately left of B, A lives two floors above B), unary attribute (A owns a Red car), and global cardinality (exactly two persons live on floor 3). Solving such hybrids efficiently requires a systematic propagation order: first apply all unary constraints to prune domains, then enforce binary relational constraints using arc consistency (AC‑3 algorithm) — repeatedly remove values from a variable’s domain that have no support in the related variable’s domain. After arc consistency stabilises, apply global cardinality constraints by counting remaining feasible assignments; if a floor’s capacity is exceeded, backtrack. A practical hand‑solved technique mimics this: draw a master grid with persons as rows and attribute categories as columns. Fill in direct clues immediately. For each relational clue, write a provisional link (e.g., “A – left – B”) and update the grid when a position becomes forced. Use “pencil marks” for multiple possibilities and eliminate them as other clues resolve. A common twist is a “cross‑category” clue such as “The person who lives on the top floor drives a Blue car”. This couples the floor variable with the colour variable, creating a new binary constraint that must be propagated across both dimensions. Ignoring cross‑category links leads to multiple apparently valid grids, whereas the correct grid is unique only after all cross‑links are honoured. Proficiency develops through practising puzzles with at least three intertwined categories and timing the propagation steps.

15. Direction Sense with Variable Speed and Non‑Orthogonal Turns

Standard direction problems assume unit steps and right‑angle turns. Advanced variants introduce variable step lengths (e.g., “walk 2 × the previous distance”) and turns of 45°, 60°, or 120°. The vector‑addition framework still applies: each leg is a vector **v**ᵢ = dᵢ · **u**ᵢ where dᵢ is the scalar distance and **u**ᵢ is a unit direction vector. For orthogonal grids, **u**ᵢ ∈ {(1,0), (0,1), (−1,0), (0,−1)}. For a 45° turn, the new heading is obtained by rotating the current unit vector by ±45° using the rotation matrix (cos θ, −sin θ; sin θ, cos θ). Because cos 45° = sin 45° = √2/2, the resulting components are rational multiples of √2. When distances are expressed as multiples of a base variable a, the final coordinate pair becomes (a·p, a·q) where p, q are algebraic numbers involving √2. The net displacement magnitude is a·√(p²+q²); the direction bearing is arctan(q/p). If the question asks only for the final bearing, the factor a cancels, leaving a pure angle. A common error is to treat a 45° turn as a simple “right” or “left” on the orthogonal grid, which yields an incorrect heading. Another variant: the path may be described from the perspective of a person facing South; left/right are reversed relative to the fixed North‑up frame. Always re‑orient the narrative to a fixed reference before vectorising. For non‑orthogonal lattices (hexagonal, triangular), define basis vectors **b**₁, **b**₂ that span the lattice (e.g., **b**₁ = (1,0), **b**₂ = (½, √3/2) for a hexagonal grid). Express each step as an integer combination of the basis vectors; the final lattice coordinate is the sum of those combinations. The Euclidean distance is then computed from the Cartesian representation of the lattice coordinate. Practising with a mix of orthogonal, 45°, and hexagonal grids builds the flexibility to recognise the underlying vector algebra regardless of the surface description.

  • Assuming a single relation label across an analogy chain without verifying each link.
  • Drawing a conclusion from a syllogistic network that contains a hidden contradiction.
  • Neglecting cross‑category constraints in hybrid puzzles, producing multiple seemingly valid grids.
  • Treating a 45° or 60° turn as a right‑angle turn in direction problems.
  • Failing to re‑orient left/right when the narrative perspective is not North‑facing.
  • Overlooking the cancellation of a common distance variable when only bearing is required.
  • Applying arc consistency only once instead of iterating until a fixed point is reached.

16. Propositional Logic Foundations for Syllogistic Extension

Classical syllogisms operate on categorical propositions (A, E, I, O) that assert inclusion or exclusion between classes. Modern CSAT items frequently embed these categorical statements inside compound sentences joined by logical connectives — conjunction (and), disjunction (or), implication (if…then), and biconditional (if and only if). Recognising the truth‑functional behaviour of each connective allows the test‑taker to translate a verbose premise into a formal structure that can be manipulated with Venn diagrams or truth tables.

A conjunction P ∧ Q is true only when both components are true; a disjunction P ∨ Q is true when at least one component is true. The material implication P → Q is false only in the case P true and Q false — a point that often traps candidates who read “if P then Q” as a causal claim. The biconditional P ↔ Q asserts equivalence and is true exactly when P and Q share the same truth value. When a premise reads “Either all managers are graduates or some engineers are not graduates”, the logical form is (∀x Mx → Gx) ∨ (∃x Ex ∧ ¬Gx). The disjunction distributes over the quantifiers only under strict conditions; the safe approach is to treat each disjunct as an independent categorical statement and test each against the conclusion.

Negation interacts with quantifiers via the square of opposition: ¬∀x (Px → Qx) is equivalent to ∃x (Px ∧ ¬Qx), and ¬∃x (Px ∧ Qx) is equivalent to ∀x (Px → ¬Qx). Mastery of these equivalences lets you convert a negative premise into a positive one, which is easier to diagram. For instance, “No student who missed the deadline passed” becomes “All students who missed the deadline are non‑passers”, an A‑type proposition ready for Venn shading.

When multiple premises contain mixed connectives, construct a truth‑table for the propositional skeleton first, then instantiate the categorical content for each satisfying row. This two‑stage method prevents the common error of mixing propositional and categorical reasoning in a single diagram. Practice with at least three premises per set builds the habit of isolating the connective layer before applying syllogistic rules.

17. Analogy Taxonomy: Semantic, Structural, and Functional Isomorphisms

Analogies in competitive reasoning are not limited to simple word pairs; they appear as triplets, quadruplets, and even matrix‑style grids where each cell must preserve a relational pattern. A robust taxonomy distinguishes three isomorphism levels. Semantic isomorphism preserves meaning relations such as synonymy, antonymy, hyponymy, or meronymy (part‑whole). Example: “thermometer : temperature :: barometer : pressure” maps instrument to measured quantity.

Structural isomorphism preserves the internal architecture of the terms — letter patterns, phonological structure, or morphological composition. “listen : silent :: post : stop” relies on anagram rearrangement; “unhappy : happiness :: unfair : fairness” exploits derivational morphology (prefix‑root‑suffix). Recognising structural cues requires scanning for repeated substrings, reversed order, or consistent affix changes.

Functional isomorphism preserves the role a term plays in a system or process. “key : lock :: password : account” maps an access credential to the secured entity; “catalyst : reaction :: enzyme : metabolism” maps an accelerator to the process it speeds. Functional analogies often appear in science‑themed sets where domain knowledge helps, but the relational skeleton remains the same across domains.

Advanced items blend levels: a first pair may be semantic, the second structural, and the third functional, yet the overall pattern demands a single mapping rule that works across all three. The solution strategy is to hypothesise the rule from the first two pairs, test it on the third, and discard if it fails. When multiple rules fit the first two pairs, the third pair acts as a discriminator. Practising with deliberately mixed‑level sets trains the mind to hold several candidate rules simultaneously and to eliminate them systematically.

Worked Analogy Chain
Complete the analogy set: “architect : blueprint :: composer : score :: programmer : ?” Options: (a) code (b) algorithm (c) syntax (d) debug
Identify the relation in the first pair: an architect creates a blueprint, a plan that guides construction. The second pair: a composer creates a score, a plan that guides performance. The relation is “creator : authoritative plan”. Among the options, a programmer creates code, which is the authoritative plan for software execution. Algorithm is a conceptual method, not the concrete plan; syntax is a rule set; debug is an action. Therefore the correct answer is (a) code.
Exam‑Day Rule for Analogy Chains
Write the relation phrase in words before scanning options. If the phrase fits more than one option, check the next pair for a tighter constraint. Never guess on the first plausible match; the test often includes a distractor that satisfies a superficial relation.

18. Circular Seating with Facing Directions and Variable Gaps

Circular arrangement problems become substantially harder when persons may face either the centre or outward, and when the number of seats between two specified persons is given as a range rather than a fixed count. The facing direction flips the meaning of “immediate left” and “immediate right”: for a centre‑facing person, left is clockwise; for an outward‑facing person, left is anticlockwise. A systematic approach is to assign each seat a binary variable Fᵢ ∈ {C, O} (centre, outward) and a positional index modulo N (total seats).

Clues of the type “A sits third to the left of B” translate to a modular equation: if B faces centre, position(A) ≡ position(B) − 3 (mod N); if B faces outward, position(A) ≡ position(B) + 3 (mod N). When the facing of B is unknown, the clue yields two possible equations, branching the search space. Variable‑gap clues such as “there are either two or three persons between C and D” produce a disjunction of modular constraints: |pos(C) − pos(D)| ∈ {3, 4} (mod N).

To solve efficiently, first list all persons with known facing (often given directly). Propagate their left/right definitions to neighbours. Next, apply fixed‑gap clues to reduce the domain of each person’s seat index. For each variable‑gap clue, keep both possibilities as parallel branches; use a depth‑first search with backtracking, but prune aggressively by checking global constraints — e.g., no two persons can occupy the same seat, and the total number of centre‑facing persons may be fixed by a clue like “exactly four face the centre”.

A practical hand‑solved technique: draw two concentric circles, inner for centre‑facing, outer for outward‑facing. Place persons on the appropriate circle as clues fix their facing. Use pencil marks for uncertain seats. When a branch leads to a contradiction (duplicate seat or violated cardinality), erase that branch and continue. This visual‑branch method mirrors the algorithmic backtracking but remains manageable for N ≤ 12.

19. Blood Relations: Multi‑Generational Trees with Adoption and Half‑Siblings

Standard blood‑relation puzzles assume biological parent‑child links and full siblings. Advanced versions introduce adoption, step‑relations, and half‑siblings, which break the simple “parent of” transitivity. An adopted child is legally a child of the adoptive parents but biologically unrelated; a half‑sibling shares exactly one biological parent. Represent each person as a node with two parental edges: bio‑father, bio‑mother, and two legal edges: legal‑father, legal‑mother. A clue such as “X is the adopted son of Y” creates a legal‑father or legal‑mother edge from Y to X without a biological edge.

When a question asks “How is P related to Q?”, the answer must specify the relation type (biological, legal, step, half) if the puzzle provides enough information to distinguish. If the data are insufficient, the correct response is “cannot be determined”. A common trap is to assume that “brother” always means full brother; the presence of a half‑sibling clue forces the solver to keep both possibilities open until a later clue eliminates one.

To diagram, draw a directed graph with solid arrows for biological links and dashed arrows for legal links. Label each arrow with the relation (father, mother, son, daughter). For half‑siblings, draw a shared biological parent node with two child nodes. Step‑relations appear as a legal parent linked to a child who already has a different biological parent. Propagation rules: legal parent of a legal child is a legal grandparent; biological parent of a legal child is not automatically a legal grandparent unless adoption is by both spouses.

Practice with at least three generations and mixed relation types. The key discipline is to never collapse legal and biological edges unless a clue explicitly states “legally and biologically”. Maintaining the distinction throughout the diagram prevents the most frequent error: assigning a biological relationship where only a legal one exists.

20. Direction Sense: Relative Motion and Vector Addition

Classic direction problems assume a fixed global north. Advanced variants describe movement relative to a moving reference such as “the person walks 5 m north relative to the train, which itself moves east at 10 m/s” or place the observer on a rotating platform. The solution requires vector addition of the reference frame’s motion and the person’s relative motion. Let the platform’s angular velocity be ω (rad/s) and the person’s velocity relative to the platform be **v**ᵣ. The absolute velocity **v**ₐ = **v**ᵣ + ω × **r**, where **r** is the position vector from the rotation centre. For a person walking radially outward on a clockwise‑rotating disc, the cross‑product term deflects the apparent path.

When only the final bearing is asked, the magnitude of ω and the time of travel often cancel if the path is described in terms of relative angles. Example: “A man stands on a rotating platform facing the centre. He walks straight ahead 10 m relative to the platform while the platform rotates 90° clockwise. What is his final direction relative to the fixed ground?” Represent the initial facing as a unit vector **u**₀. After rotation, the platform’s axes have turned by 90°, so the man’s relative displacement vector in ground coordinates is the rotation of **u**₀ by 90°. His final bearing is therefore the initial bearing plus 90°.

For problems with variable speed, express each leg as a vector **v**ᵢ = dᵢ **u**ᵢ where dᵢ may be a function of previous distances (e.g., d₂ = 2 d₁). Sum the vectors component‑wise. If the question asks only for the angle of the resultant, factor out the common distance variable and compute arctan of the ratio of summed components. This cancellation is a frequent time‑saver.

Non‑orthogonal turns (45°, 60°, 120°) are handled by rotating the unit vector using the rotation matrix with cos and sin of the turn angle. Because cos 45° = sin 45° = √2/2, the components become multiples of √2. Keep the √2 factor symbolic; it cancels in bearing calculations. Practise with a mix of rotating platforms, moving vehicles, and non‑orthogonal turns to internalise the vector‑addition workflow.

  • Confusing legal and biological edges in multi‑generational family trees.
  • Treating “left” as clockwise for all persons in circular seating regardless of facing direction.
  • Assuming a single analogy rule applies across all pairs without testing the third pair.
  • Neglecting the cross‑product term when the reference frame rotates, leading to a bearing error.
  • Failing to cancel a common distance variable when only the final bearing is required.
  • Review each trap before the exam to avoid losing marks.

21. Conditional Reasoning: Nested Implications and Contrapositive Chains

Conditional reasoning extends the simple “if‑then” structure into layers where the consequent of one implication becomes the antecedent of another. A nested chain such as P → (Q → R) is logically equivalent to (P ∧ Q) → R by exportation, but tests often require tracking the contrapositive at each level. The contrapositive of P → Q is ¬Q → ¬P; for a nested form the contrapositive propagates inward: ¬R → ¬(Q → P) which simplifies to ¬R → (Q ∧ ¬P). Mastery of these transformations lets a candidate eliminate answer choices that only appear plausible under a surface reading.

Two mechanical tools are indispensable. First, a truth‑table sketch for three variables (8 rows) confirms equivalence without memorising every derived rule. Second, a diagrammatic “implication tree” where each node is a proposition and arrows point from antecedent to consequent. When a premise states “If A then (if B then C)”, draw A → (B → C). To test a conclusion such as “If not C then not A”, trace the tree backwards: ¬C forces ¬(B → C) which forces B ∧ ¬C; this does not directly give ¬A unless B is known true. Wording such as “provided that” or “whenever” often hides the extra conjunct B. Recognising that the hidden conjunct must be affirmed before the contrapositive yields ¬A is the decisive step.

Common patterns include: (i) a chain of three conditionals where the middle term is a disjunction, e.g. P → (Q ∨ R) and Q → S, R → S; the valid inference is P → S by constructive dilemma. (ii) A biconditional disguised as two conditionals with a shared term, e.g. “A only if B” and “B only if A” together give A ↔ B. (iii) A conditional with a negative antecedent, “Unless A, B” which translates to ¬A → B and its contrapositive ¬B → A. Practise converting each English cue into symbolic form before drawing the tree; this eliminates the linguistic ambiguity that tests exploit.

22. Logical Games: Multi‑Variable Ordering with Conditional Constraints

Advanced ordering games present several independent sequences — such as floors, days, and colours — that must be aligned simultaneously. Each entity carries a tuple of attributes (person, floor, colour, hobby). Constraints are of three types: absolute (e.g., “A lives on floor 3”), relative (e.g., “B is two floors above C”), and conditional (e.g., “If D likes red then E lives on an even floor”). The solution space is a Cartesian product of the individual domains; a systematic approach reduces it by constraint propagation.

Begin by constructing a master grid: rows = entities, columns = attribute categories. Fill in absolute clues first. Next, encode relative clues as inequalities between row indices (e.g., floor(B) = floor(C) + 2). Conditional clues are stored as implication rules attached to the relevant cells. Propagation proceeds iteratively: whenever a cell receives a single possible value, eliminate that value from all other cells in the same column (the “unique‑value” rule) and fire any conditional rule whose antecedent becomes true. If a conditional’s antecedent becomes false, the rule is dormant; if undetermined, the rule remains pending.

A powerful shortcut is the “forced‑pair” technique. When two entities share the same set of two possible floors and no other entity can occupy those floors, the pair must occupy them in some order. This creates a binary sub‑game solvable by trying both permutations; only one will satisfy all pending conditionals. Another technique is “domain splitting”: pick an entity with the smallest domain (≥2 values), branch on each value, and propagate. Because the total number of entities is typically small (≤ 8), the search tree remains shallow. Record each branch’s deductions in a separate copy of the grid; the branch that reaches a contradiction is discarded. The surviving branch yields the unique arrangement.

Worked Example: Conditional Ordering
Five friends — P, Q, R, S, T — live on floors 1‑5 (one per floor). Each likes a different colour: Red, Blue, Green, Yellow, White. Clues: (1) P lives on floor 3. (2) Q likes Blue. (3) The person who likes Red lives immediately above the person who likes Green. (4) If S likes Yellow then T lives on floor 5. (5) R does not like White. Determine the floor and colour of each person.
Step 1: Place P on floor 3. Step 2: Q‑Blue placed later. Step 3: Red‑Green adjacency means Red floor = Green floor + 1. Possible floor pairs: (2,1), (3,2), (4,3), (5,4). Floor 3 is occupied by P, so (3,2) and (4,3) are impossible because Green would be on floor 2 or 3 but floor 3 is P (colour unknown). Remaining pairs: (2,1) and (5,4). Step 4: Conditional (4) – if S=Yellow then T=5. Keep pending. Step 5: R≠White. Step 6: Try Red‑Green = (2,1). Then floors 1=Green, 2=Red. Floors left for colours: 4,5 for Yellow, White, Blue. Q=Blue must occupy one of 4 or 5. S=Yellow would force T=5 (clue 4). If S=Yellow on floor 4, T=5 works; remaining floor 5 gets White, but R≠White so R cannot be on floor 5. Assign R to floor 4 (Yellow) – conflict because S=Yellow. So S≠Yellow. Then Yellow goes to floor 5, T not forced. But then T would be on floor 4 or 5; check consistency – leads to multiple solutions. Step 7: Try Red‑Green = (5,4). Floors 4=Green, 5=Red. Remaining floors 1,2 for Yellow, White, Blue. Q=Blue on 1 or 2. Conditional (4): if S=Yellow then T=5, but floor 5 already Red, so T cannot be 5 → antecedent false → S≠Yellow. Thus Yellow on floor 1 or 2. R≠White. Exhausting permutations yields unique assignment: Floor 1 = S‑Yellow, Floor 2 = Q‑Blue, Floor 3 = P‑White, Floor 4 = R‑Green, Floor 5 = T‑Red. All clues satisfied.
Exam‑Day Strategy
When a game contains conditional clues, write each as an explicit “If X then Y” on a side sheet. During propagation, the moment a conditional’s antecedent becomes true, immediately apply its consequent; if the antecedent becomes false, cross the rule out. This prevents the common mistake of revisiting a dormant rule later and re‑activating it incorrectly.

23. Data Sufficiency: Minimal Information Principle and Decision Trees

Data sufficiency questions ask whether the given statements, individually or together, provide enough information to answer a target question. The minimal information principle states that a statement is sufficient only if it eliminates every possible world except those that agree on the answer. To apply this, model the problem as a set of possible assignments (worlds) consistent with the premise. Each statement partitions this set; a statement is sufficient if the partition cell containing the actual world is a subset of the answer‑yes worlds or answer‑no worlds exclusively.

Construct a decision tree: the root represents the initial uncertainty. Branch on Statement 1 (true/false/insufficient). If a branch yields a unique answer, mark it sufficient. Otherwise, continue branching on Statement 2 within that branch. The combined sufficiency corresponds to the leaf reached after both statements. A practical shortcut is to test extreme cases: for a numeric question, try the minimum and maximum values allowed by each statement. If both extremes give the same answer, the statement is sufficient; if they differ, it is not. For relational questions (e.g., “Is A taller than B?”), construct two concrete scenarios that satisfy the statement but give opposite answers. The existence of such a pair proves insufficiency.

Typical traps: (i) a statement that appears to give a numeric value but actually leaves a hidden degree of freedom (e.g., “x + y = 10” does not fix x). (ii) Two statements that together still leave a binary ambiguity because they constrain the same variable in the same way. (iii) Over‑reliance on algebraic manipulation without checking whether the derived expression still contains an unknown. Always verify that the final expression depends solely on known constants.

24. Argument Mapping: Identifying Hidden Premises and Formal Fallacies

Critical reasoning passages often hide premises necessary for the conclusion to follow. Argument mapping makes the structure visible: write the conclusion at the top, draw arrows from each explicit premise, then ask “What else must be true for this inference to be valid?” The missing links are the hidden premises. For example, “All managers are leaders. Therefore, some leaders are innovative.” The hidden premise is “Some managers are innovative.” Without it, the conclusion does not follow. Mapping also reveals formal fallacies: affirming the consequent (If P then Q; Q; therefore P), denying the antecedent (If P then Q; not P; therefore not Q), and the fallacy of the undistributed middle in categorical syllogisms.

To detect the undistributed middle, translate each categorical statement into standard form (All A are B, Some A are B, No A are B, Some A are not B). Identify the middle term — the term appearing in both premises but not in the conclusion. Check whether the middle term is distributed (refers to all members of its category) in at least one premise. If not, the syllogism commits the undistributed middle fallacy. Another frequent pattern is the illicit major/minor: the major or minor term is distributed in the conclusion but not in its premise. Mark distribution with a superscript “d” on the term; the rule “no term may be distributed in the conclusion unless distributed in the premise” catches both errors instantly.

Practice by rewriting every argument in the passage into a numbered list of premises and a conclusion, then apply the distribution checklist. Tests often present a “strengthen/weaken” question; the correct answer is the statement that either supplies a missing premise (strengthen) or introduces a counter‑example to a hidden premise (weaken). Recognising the hidden premise first turns these questions into a direct lookup.

25. Temporal Reasoning: Partial Order Scheduling and Critical Path Extraction

Temporal reasoning problems give a set of events with precedence constraints (e.g., “A occurs before B”, “C and D cannot be simultaneous”). The goal is to determine possible orders, earliest/latest times, or whether a specific ordering is forced. Represent each event as a node in a directed acyclic graph (DAG); an edge X → Y means X must precede Y. The partial order is the transitive closure of these edges. A topological sort yields all linear extensions; the number of extensions can be large, but questions typically ask for properties common to all extensions (necessary precedences) or for the existence of an extension satisfying an extra condition.

The critical path method (CPM) adapts to discrete events by assigning each event a duration (often 1 unit). The earliest start time (EST) of a node is the maximum EST of its predecessors plus their durations. The latest start time (LST) is computed backward from a designated finish node. Slack = LST − EST; events with zero slack lie on every critical path and their relative order is fixed. If a question asks “Which event must occur before event X?”, compute the set of ancestors of X in the DAG; any ancestor with zero slack is a necessary predecessor. Conversely, if two events are incomparable (no directed path either way), their order can be swapped in some valid schedule.

Advanced variants introduce “resource constraints” (only one event per time slot) turning the problem into a scheduling puzzle. The solution then combines topological ordering with a greedy slot‑filling algorithm: at each time step, schedule any available event (all predecessors already scheduled) preferring the one with the smallest LST (earliest deadline first). This heuristic guarantees a feasible schedule if one exists for unit‑duration tasks. For exams, it suffices to simulate the first few steps manually; answer choices usually hinge on whether a particular event can be placed at a given slot.

  • Assuming a conditional’s antecedent is true simply because the consequent appears in the passage.
  • Treating a conditional ordering clue as bidirectional when only one direction is stated.
  • Overlooking a hidden degree of freedom in data‑sufficiency statements that leaves the answer ambiguous.
  • Failing to check distribution of the middle term before accepting a categorical syllogism as valid.
  • Ignoring slack values and assuming all ancestors of a node are necessary predecessors.
  • Applying the earliest‑deadline‑first rule without verifying that all predecessors are already scheduled.
  • Confusing “unless” with “if not” and reversing the contrapositive direction.

Practice Questions

← Reading Comprehension Quantitative Aptitude →