Non-Verbal Reasoning for the 11+: Question Types, Techniques and Free Practice

What Is Non-Verbal Reasoning?

Non-verbal reasoning (NVR) is the ability to analyse visual information, identify patterns, and solve problems using shapes, diagrams, and spatial relationships — without relying on written or spoken language. It is a measure of fluid intelligence: the capacity to reason through novel problems rather than recall learned knowledge.

NVR is included in the 11+ because it offers a more equitable measure of cognitive potential than verbal or mathematical tests alone. A child from a home where English is not the first language, or one who has had less exposure to books and formal academic preparation, is not disadvantaged in NVR in the same way they might be in a verbal reasoning test. It levels the playing field — or at least tilts it less steeply.

The 11+ NVR paper is used by both major exam boards: GL Assessment and CEM. GL Assessment tends to use a dedicated NVR paper as part of its multi-paper format. CEM integrates NVR questions within a mixed reasoning paper. Both formats assess the same underlying skills, but the experience of the test — and the weighting of NVR in the final score — differs between exam boards.

The Six Question Types

There are six main question types in 11+ NVR. Each type follows its own internal logic and responds best to a specific technique. Children who learn the technique for each type — rather than approaching every question by intuition — improve faster and make fewer errors under time pressure.

1. Figure Matrices

A 2×2 or 3×3 grid of shapes with one cell missing. The task is to identify the missing shape from a set of options. The grid follows a consistent rule — each row and column relates to the others in a specific way.

• Identify the rule operating in the top row first: is the shape rotating? Gaining or losing elements? Changing shading? Changing size?
• Check that the same rule operates in the second row. If it does, the rule is confirmed.
• Apply the confirmed rule to complete the missing cell, then verify your answer also satisfies the column relationships.
• If two rules seem possible, test both against all rows and columns — only one will be consistent throughout the whole grid.

2. Figure Series

A sequence of four or five shapes in which each shape relates to the next by a consistent rule. The task is to identify what comes next in the sequence.

• Look for ONE change between each consecutive pair of shapes rather than trying to see the whole pattern simultaneously.
• Common changes: rotation (clockwise or anticlockwise by a consistent number of degrees), addition or subtraction of elements, change in shading, change in size.
• Once you identify the change between shapes 1 and 2, check it holds between shapes 2 and 3, then 3 and 4. If it does, apply it to predict shape 5.
• Watch for alternating patterns — the rule may apply to every other shape rather than every consecutive pair.

3. Figure Analogies

Shape A relates to shape B in a specific way. Shape C is given; the task is to find shape D that relates to C in the same way that B relates to A.

• Describe the transformation from A to B in plain language before looking at the answer options: “It rotated 90 degrees clockwise, and the shading changed from black to white.”
• Apply that exact transformation to shape C. The answer should follow mechanically from your description.
• If your predicted answer does not appear in the options, revisit your description of the A→B transformation — you have missed an element of the rule.

4. Figure Classification (Odd One Out)

Five shapes are shown. Four share a common property; one does not. The task is to identify the odd one out.

• Do not try to identify the odd one out by intuition. Instead, find what the majority (four shapes) have in common.
• Check systematically: number of sides, presence and type of shading, orientation, inner elements, symmetry.
• The common property must apply to exactly four shapes and exclude exactly one.
• If you find a property shared by only three shapes, keep looking — the grouping must be four, not three.

5. Figure Codes

Each of three or four shapes is paired with a two-letter code. A new shape is given; the task is to identify its code from the options provided. Each letter in the code corresponds to one property of the shape.

• Decode each letter separately. Never try to decode both letters simultaneously.
• For the first letter: look at what all shapes sharing the same first letter have in common. This is what the first letter encodes (usually the outer shape or basic form).
• For the second letter: look at what shapes sharing the same second letter have in common (usually shading, inner element, or pattern).
• Apply both decoded rules to the new shape to construct its code. Never guess figure codes — always decode systematically.

6. Spatial Rotation

A shape is shown alongside three or four answer options, some of which are rotations of the original and some of which are reflections or entirely different shapes. The task is to identify the correctly rotated version.

• Identify one distinctive feature of the original shape — a notch, a protruding element, a dot, an asymmetry — before looking at the options.
• Track where that distinctive feature moves under the stated rotation. Use this as your anchor point rather than trying to visualise the whole shape rotating.
• Eliminate options where the distinctive feature is in the wrong position. Usually only one option survives.
• Remember that rotation preserves the shape; reflection reverses it. If an answer option looks like a mirror image, it is a reflection — not a rotation — and should be eliminated.

Common Mistakes and How to Fix Them

• Rushing figure matrices without identifying the rule first — leads to guessing from the options rather than predicting the answer. Fix: always articulate the rule in words before consulting the answer options. If you cannot state the rule, you have not found it yet.
• Confusing rotation with reflection — these look similar when a shape has partial symmetry. Fix: practise with physical objects; hold a shape up to a mirror and observe the difference between what you see in the mirror and what you see when you rotate the object. The distinction becomes intuitive with practice.
• Spending too long on figure series questions — series questions that require tracking multiple simultaneous changes can be very time-consuming. Fix: if you have not identified the rule within approximately 30 seconds, mark the question and return to it at the end. Running out of time on easy questions because you were stuck on a hard one is a preventable error.
• Guessing figure codes — more than any other question type, codes reward systematic decoding and punish guessing. A child who guesses codes will score at chance (20% with five options). Fix: drill the decode-each-letter-separately technique until it is automatic, even under time pressure.
• Not verifying the answer against all options — finding an answer that seems to fit and immediately marking it, without checking whether another option fits better. Fix: always apply your rule to your chosen answer and confirm it is consistent before moving on. This takes five seconds and prevents a significant fraction of errors.

GL Assessment vs CEM: What's the Difference?

The two major 11+ exam boards in England handle non-verbal reasoning differently, and knowing which your area uses is important before you invest in practice materials.

GL Assessment — used by grammar schools in Kent, Buckinghamshire, Hertfordshire, Lincolnshire, and most other selective areas — typically uses a dedicated NVR paper as part of a multi-paper format. The NVR paper contains questions across all six types described above, with clear section breaks. The time per question is predictable and preparation materials are widely available. If your area uses GL Assessment, purchasing dedicated GL NVR practice papers is straightforward and effective.

CEM (Centre for Evaluation and Monitoring) — used by some grammar schools in Birmingham, Durham, and parts of the North East — integrates NVR questions within a mixed verbal and non-verbal paper, with no advance warning of which type of question comes next. This means children need to be comfortable switching between verbal and non-verbal tasks rapidly. CEM also tends to use a narrower range of NVR question types than GL, with matrices and series featuring most prominently. Check with your target school or local grammar school association which exam board they use before selecting practice materials.

How to Practise Effectively

The most common preparation mistake is treating all six NVR question types as equally important and practising them in equal proportion. They are not equally important — and, more relevantly, your child is almost certainly stronger in some types than others. Effective preparation starts with a diagnostic.

• Start with a diagnostic assessment: Before buying workbooks or starting a practice schedule, find out which question types your child finds hardest. A good adaptive tool (like Eduentry) will surface this automatically. A manual diagnostic can be done by working through one set of each question type and noting accuracy and time per type.
• Prioritise adaptive practice over fixed-difficulty papers: Fixed-difficulty practice papers quickly become too easy as a child improves — they stop producing the cognitive challenge needed for continued gains. Adaptive tools adjust difficulty in real time, keeping your child in the productive zone of challenge. Eduentry adjusts NVR question difficulty automatically and reports scores on the same standardised age score (SAS) scale as GL Assessment.
• Short daily sessions outperform long weekend sessions: Twenty to thirty minutes of focused NVR practice five days per week produces better results than a two-hour session on Saturday. Cognitive skill consolidation happens between practice sessions, not just during them. Spacing matters.
• Use physical and everyday activities to build spatial skills:Tangrams, Rubik's cube, tessellation puzzles, origami, and construction toys all develop the spatial reasoning that underlies NVR performance. These are not substitutes for test practice, but they build the foundational capacity that makes test practice effective.
• Track scores over time and expect gradual improvement: NVR gains are rarely sudden. Expect measurable improvement within four to six weeks of consistent practice, with significant gains over three to six months. If scores are not improving after six weeks of consistent practice, the question difficulty may be pitched incorrectly — either too easy (not challenging enough to drive improvement) or too hard (producing frustration and guessing rather than learning).

What Score Do You Need?

Most 11+ NVR papers are reported as standardised age scores (SAS) — a scale where 100 is the population mean for age and each 15 points represents one standard deviation. This allows direct comparison between children of different ages, and between scores across different test sittings.

Grammar school entry benchmarks, as a general guide:

• SAS 111+ — competitive for most grammar schools; approximately the 77th percentile for age
• SAS 115 — approximately the 84th percentile; comfortably above threshold for most grammar schools
• SAS 120 — approximately the 91st percentile; competitive for selective grammar schools
• SAS 130+ — approximately the 98th percentile; highly competitive for the most selective schools

It is important to note that most 11+ systems use a combined score across all papers — verbal reasoning, non-verbal reasoning, maths, and (in some areas) English. A strong NVR score can compensate for a weaker verbal score in systems that average across all components. Knowing your child's relative strengths across all domains — not just NVR — allows you to identify where targeted effort will produce the biggest improvement in their overall profile.

Conclusion

Non-verbal reasoning rewards systematic preparation in a way that few other academic skills do. Unlike English or mathematics, it does not require extensive background knowledge or years of curriculum study. What it requires is trained pattern recognition — knowing the six question types, applying the right technique for each, and practising consistently at the right level of difficulty.

The practical framework is straightforward: measure first (use a standardised benchmark to identify which question types are weakest), focus preparation on those types (not the ones your child already finds easy), practise adaptively (ensure difficulty tracks your child's improving level), and track progress (expect meaningful improvement within six weeks of consistent practice).

The children who sit the 11+ and find the NVR paper manageable are not more gifted — they are better prepared. That preparation is available to any child who starts early enough and practises with the right approach.