Non-verbal Pattern Completion

Overview

Non-verbal Pattern Completion appears about 1.8 times per paper across the last four AFCAT solved papers, placing it in the high yield band of Reasoning and Aptitude.

Non-verbal pattern completion is the AFCAT question where you see a small figure grid — usually a 2x2 or 3x3 layout — with one cell left blank, and you choose the answer figure that completes the design. About 1.75 marks per paper hang on it, sometimes spread between a pure grid item and a series-of-figures item where four shapes sit in a row and you supply the fifth.

The trick is that the figures themselves are simple — squares, circles, triangles, arrows, dots, line segments — but the rule that ties them together is hidden. You have to scan rows, scan columns, sometimes scan diagonals, and pin down a single transformation: a rotation, a reflection, an added element, a shading change. Once the rule is named in plain words, the answer figure picks itself, and the three wrong options become obvious because each one breaks the rule in a different way.

This page works through the question shape, the grid-scan method, the full transformation catalogue, the way AFCAT lays traps, and ten worked examples written in clean text descriptions so you can follow them without printed figures.

Why non-verbal items reward visual fluency

Verbal reasoning runs on words; non-verbal reasoning runs on shape memory. The pattern-completion item gives you between thirty seconds and a minute to compare up to eight tiny pictures, name the rule and pick from four answer figures. There is no formula to write. The only resource is your eye and how quickly it can hold one figure in working memory while comparing it against the next.

Visual fluency is trainable, and that is what makes this question paper-friendly. A student who has drilled twenty figure grids can name a 90-degree clockwise rotation as fast as a chess player spots a pin. A student who has never practised stares at the grid waiting for it to speak. The marks per minute on this topic are excellent once the eye is trained — most candidates who have done a few hundred items solve a fresh one in under forty seconds at near-perfect accuracy.

The other reward is psychological. Pattern items sit alongside heavier algebra and reading-comprehension questions in the AFCAT paper. Knocking off a non-verbal item in under a minute lifts confidence and keeps the rhythm of the paper steady. Skipping it because the figures look unfamiliar is a small but real morale loss.

Question shape — the partial grid

The most common AFCAT format shows a 3x3 grid of nine cells. Eight cells contain small figures; the ninth cell — usually the bottom-right but sometimes a middle cell or the top-right — is blank, replaced by a question mark. Below the grid sit four answer figures labelled (A), (B), (C), (D). Your job is to pick the one figure that fills the blank cell so that every row and every column obeys a consistent rule.

A second common format is the 2x2 grid. Three cells are filled; one is blank. The rule is usually simpler — only one row pair and one column pair to scan — but the answer figures are designed to look almost identical, so attention to detail matters more.

A third format, called the series-of-figures variant, drops the grid altogether. Four shapes sit in a horizontal row with a question mark in the fifth position. The rule is applied from each figure to the next. Whichever shape the rule lands on at position five is the answer.

In all three shapes the figures are intentionally minimalist — outlines, dots, shading, simple arrows. The simplicity hides the rule.

The grid-scan method — rows, columns, diagonals

Use a fixed scan order so you never miss a rule.

1. Scan the rows. Take row 1, name how figure 1 changes into figure 2, then check whether the same change carries from figure 2 to figure 3. If it does, you have the row rule. Confirm by running the rule across row 2.
2. Scan the columns. Take column 1 and ask the same question top to bottom. If the row rule has fixed the missing cell, the column scan should agree — if it does not, you have misread one of the rules.
3. Scan the diagonals. Some grids hide the rule along the main diagonal (top-left to bottom-right) or the anti-diagonal. Always glance at both before committing.
4. Predict in words before looking at answer figures. If your prediction is "a triangle with shading on the right, pointing down," you will reject distractors instantly. If you look at the answer figures first you will be biased by the most attractive one.
5. Match and eliminate. Reject options that break the row rule, then reject options that break the column rule. The survivor is the answer.

The full scan takes around fifteen seconds when practised. The remaining time is for confirmation.

Common transformation rules

Almost every AFCAT pattern grid uses one of these eight rules, applied in isolation across a row or column.

Keep this table in mind. When you read a row you should be able to silently tick which rule is in play within five or six seconds.

Rotation directions — clockwise versus counter-clockwise

Many candidates lose marks not on the angle but on the direction of rotation. The clock face is the anchor. Picture twelve at the top, three on the right, six at the bottom, nine on the left. Clockwise is twelve to three to six to nine. Counter-clockwise is the reverse.

Apply this to any figure with a distinguishing tip — an arrow, a dot, a flag, the longer end of a line.

To detect direction in a grid, find the figure with the most distinctive tip and trace where the tip moves from cell one to cell two. If it lands one notch towards the three-o-clock side, the rotation is 45 clockwise. If it lands at the three-o-clock side itself, the rotation is 90 clockwise. The same logic mirrored gives counter-clockwise.

The one-attribute-at-a-time rule

AFCAT pattern grids almost never change two attributes in the same step. If a square becomes a rotated square, the shading stays the same. If shading flips from hollow to solid, the rotation stays fixed. If an element is added, the existing elements do not move.

This single fact eliminates most distractors. Suppose the row rule is "rotate 90 clockwise." Among the four answer figures, you may see one that is correctly rotated but also has its shading flipped. That option is the trap. The right answer keeps shading constant and rotates only.

The exam writer plants these traps to punish candidates who lock onto the most visible transformation and miss the second invariant. Train yourself to ask, after naming the rule, "what is the rule not doing?" That second question saves at least one mark per non-verbal section.

The series-of-figures variant

Sometimes the question is not a grid at all but a horizontal row of four small figures with a question mark at position five. The transformation is applied from figure 1 to figure 2, from figure 2 to figure 3, and so on. Your task is to apply it once more.

The method is identical to the grid scan but simpler — there are no columns to worry about. Read figure 1 against figure 2, name the rule, and confirm it survives from figure 2 to figure 3 and from figure 3 to figure 4. Then apply it to figure 4 to predict figure 5.

One useful warning: in the series variant, the transformation may compound rather than repeat. A rotation of 45 may double to 90, then to 135, then to 180. If the cell-to-cell change does not look constant, check whether the change is increasing in a steady way — that itself is the rule.

Counting elements when the shape changes

When the outer shape itself changes between cells — say a triangle in cell one, a square in cell two, a pentagon in cell three — pure rotation rules cannot apply. Switch to the counting method.

1. Count the sides of each figure. If the sequence is 3, 4, 5, the next cell has six sides.
2. Count the dots, arrows or small marks inside each figure. If the count grows by one each step, the missing cell shows one more dot than the cell before.
3. Count the line segments. A figure series sometimes adds one line at each step, building a star or a wheel.

The counting method also helps with shading. If cell one has zero shaded segments, cell two has one, cell three has two, the rule is "shading increases by one segment per step." The missing cell continues the count.

Symmetry and axis check

Reflection rules are easy to miss because a horizontally reflected square looks identical to the original square. The reflection only becomes visible when the figure has an asymmetric mark — a dot in one corner, an arrow on one side, an extra line on the right.

Use this axis-check routine for any figure pair where you suspect reflection:

• Imagine a vertical line through the middle of figure one. Does figure two look like a left-right flip about that line? If yes, the rule is vertical-axis reflection.
• Imagine a horizontal line through the middle of figure one. Does figure two look like a top-bottom flip about that line? If yes, the rule is horizontal-axis reflection.
• If neither imagined line works, the rule is rotation, not reflection. A 180-degree rotation can look like a reflection at first glance — to tell them apart, check a feature in one corner. A rotation moves it to the diagonally opposite corner; a reflection moves it across the axis only.

Common AFCAT figure types

The figures in AFCAT pattern grids come from a small recurring vocabulary. Recognising the family saves time.

When you spot a family, the relevant rule narrows immediately. A polygon with shading is almost always a shading-rotation item. A circle with dots is almost always a position-shift or count item.

Trap patterns to recognise

Three common traps account for most wrong answers on this topic.

Whenever two answer figures look equally good, the trap is usually one of these five.

Practice rhythm and time budget

You will see roughly two non-verbal items per AFCAT paper. Budget about ninety seconds total — forty to fifty seconds for each. That leaves enough buffer for one re-check.

• Daily drill: Ten figure-grid items every other day for a fortnight is enough to lock the patterns into long-term memory.
• Mixed sets: Mix grid items with series-of-figures items so your eye learns to spot both formats fast.
• Annotate: When you get an item wrong, write the rule down in plain words and the trap you fell for. After two weeks the same traps will not catch you.
• Sectional mock once a week: Solve thirty mixed reasoning items in twenty-five minutes. Pattern completion sits between coding and series — keep the rhythm even.

The marginal cost of one more drill item is small; the marginal return on this topic is one of the highest in the AFCAT paper because the figure vocabulary repeats.

Worked AFCAT-style examples

Exam-day strategy

1. Always state the rule in plain words before looking at the answer figures — this stops the most attractive distractor from biasing you.
2. Run the row scan first; if it pins the answer, confirm with the column scan as a check.
3. When direction is ambiguous, trace the most distinctive tip (arrow head, flag, dot) across two cells to confirm clockwise versus counter-clockwise.
4. Apply the one-attribute-at-a-time rule — if a candidate answer changes two things, reject it.
5. For grids where the outer shape changes, switch from rotation logic to a counting check (sides, dots, line segments).
6. Watch for the 180-degree-rotation versus full-reflection trap by checking where a corner feature lands.
7. Aim for forty to fifty seconds per item; budget around ninety seconds total for the two non-verbal items in the paper.
8. Drill ten mixed figure-grid and series items every alternate day for two weeks before the exam.

Frequently asked questions

Related topics