Analogy: Reasoning Ability Notes for Dadra & Nagar Haveli and Daman & Diu Police Constable

📚 Advanced Study Notes – Analogy

Target Exam: Dadra & Nagar Haveli & Daman & Diu Police Constable – Reasoning Ability
Level: MIT‑style, from first principles to edge‑cases

Analogy is a cognitive bridge that lets you transfer knowledge from one domain (the source) to another (the target) by exposing shared relational structure.
In reasoning tests, analogies reveal your ability to detect patterns, make inferences, and generalise.

1. Foundations – From First Principles

Concept	Definition	Why It Matters
Similarity	Measure of closeness between objects/relations.	Analogy relies on similar relational patterns.
Mapping	One‑to‑one correspondence between elements of source and target.	Determines the structure you transfer.
Relation	The connective that holds elements together (e.g., “is a part of”, “causes”).	Analogies are about relations, not just attributes.
Structure‑Mapping Theory (SMT)	Cognitive model by Dedre Gentner. Analogical reasoning is the alignment of relational structures.	Provides a rigorous, computational framework.
Relational Proportion	`a : b :: c : d` – “a is to b as c is to d”.	Standard format in many exams.

2. Types of Analogies

Type	Example	Key Feature
Synonymic	Bright : Shiny	Same or similar meaning.
Antonymic	Hot : Cold	Opposite meanings.
Part‑Whole	Wheel : Car	Whole contains part.
Cause‑Effect	Fire : Smoke	One causes the other.
Function	Pen : Write	Tool and its use.
Process‑Result	Plant : Flower	Outcome of a process.
Instrument‑Action	Hammer : Nail	Tool for a specific action.
Category‑Member	Dog : Mammal	Member and category.
Analogous‑Structure (most common in reasoning tests)	River : Mountain	Both are landforms but different kinds.

3. Formal Representation

An analogy can be expressed as a relational proportion:

(a R1 b)  ::  (c R2 d)

R1 and R2 are relations that must be isomorphic (structurally identical).
a → c and b → d are mapped counterparts.

Example
(Tree : Branch) :: (Car : Wheel)
Relations: part‑of.

Checking Validity

Identify relation R.
Map each element of the source to the target.
Verify bijection (one‑to‑one).

4. Cognitive & Computational Aspects

4.1 Structure‑Mapping Theory (SMT)

Input: Source knowledge (structured facts).
Process: Mapping – align source relations to target.
Output: Transfer of consequences (inferences).

SMT explains why analogies often involve higher‑order relations (e.g., causality) rather than simple attributes.

4.2 Similarity Metrics

Metric	Formula	Use Case
Jaccard Index	`	A∩B
Cosine Similarity	`A·B / (
Edit Distance	`min{insert, delete, substitute}`	Lexical analogies.

5. Edge Cases & Common Pitfalls

Edge Case	Description	How to Handle
Partial Mapping	Only some elements match.	Look for core relational structure; ignore extraneous attributes.
Non‑bijective Mapping	Multiple elements map to one.	Impose a one‑to‑one rule for exam analogies.
Ambiguous Relations	Relation can be interpreted in multiple ways.	Use the most salient relation (e.g., part‑whole over function).
Over‑fitting	Too specific mapping (e.g., pencil → write).	Generalise to the underlying relation (tool → action).
Circular Analogies	`A : B :: B : A`	Rare; treat as symmetric relation (e.g., cause ↔ effect).

6. Practice Strategies

Identify the Relation
- Ask: What connects the two words?
- Look for verbs (uses, causes, contains).
Look for Structural Patterns
- Are they part‑whole, cause‑effect, function?
- Do they belong to the same category?