Logical / Abstract Reasoning
Abstract reasoning measures pattern detection without relying on words or numbers. You will see sequences, matrices, or odd-one-out figures and must infer the rule that produces the next item. The rule is usually simple but applied consistently.
Features to scan
- Count sides, lines, intersections, enclosed regions, and dots.
- Track position: clockwise or counterclockwise, step size, bounce at edges.
- Track orientation: rotation versus reflection.
- Track attributes: shading, size, open or closed, symmetry.
- In 3×3 matrices, test rows first, then columns; rules often combine two features.
Worked example: a square with a dot at top-left; a pentagon with a dot at top-right; a hexagon with a dot at bottom-right. The sides increase by one each step (4, 5, 6), and the dot moves one corner clockwise. The next figure is a heptagon with the dot at bottom-left. Noticing only the dot movement would miss the side count; both rules must hold.
Traps that cost marks
- Assuming rotation when the figure is reflected.
- Finding a rule that fits two items but fails the third.
- Ignoring a second attribute such as shading or size.
- Overcomplicating when a simple count works.
- Choosing an option that satisfies one rule but breaks another.