The one part of this exam that definitely responds to practice
COA says NATA measures a sensibility that cannot be coached into existence. That claim is at its weakest here. Mental rotation is one of the most thoroughly studied skills in cognitive psychology, and one of the most reliably trainable — people get measurably better at it with practice, and the improvement holds. So if you have limited drill time and want the best return on it, this is where it goes.
Stop turning the object. Turn yourself.
Most candidates attack rotation questions by trying to spin the shape in their head like a hologram. It works for simple objects and collapses for anything with more than a couple of features, because you are holding the whole object in working memory and transforming it at the same time.
The better move is to leave the object exactly where it is and move your viewpoint instead. Instead of rotate the cube 90 degrees to the right, ask what would I see if I walked round to its left side? Same geometry, dramatically less cognitive load — because now the object is fixed and only you are moving, and you have thirty years of experience walking around things.
This is not a gimmick. It is why architects tend to be good at these questions without ever having drilled them: they are used to imagining themselves inside and around a building rather than spinning buildings in the air.
Track one feature, not the whole shape
The second load-reducer. You do not need to rotate the entire object. You need to answer one question, and usually one feature decides it.
Find the most distinctive thing in the figure — the odd protrusion, the single dot, the one arm that is longer. Track only that. Where does it end up? Most answer options can be eliminated on that basis alone, and you have carried one feature through the rotation instead of twelve.
This pairs with a fact about the options: in most rotation questions, the wrong answers are wrong in specific ways. One is the mirror image. One has the right shape with a feature in the wrong place. One is rotated about the wrong axis. If you know what kind of wrong you are looking for, one tracked feature usually finds it.
Under 108 seconds, this is the difference between finishing and guessing.
Rotation never changes handedness. Ever.
This single rule kills a whole family of wrong answers, and it is worth knowing cold because it requires no visualisation at all.
A rotation cannot turn a left-handed object into a right-handed one. No amount of spinning will convert your left glove into a right glove. It is a property of the object, not of its orientation, and rotation preserves it absolutely.
So whenever a rotation question offers you an option that is a mirror image of the original, you can eliminate it instantly — not by checking, but by knowing. It is the single most common distractor in the whole area, because mirrored shapes look almost right, and a candidate who is mentally spinning the object at speed will accept one.
The practical check: find any chirality cue in the figure — a spiral's direction, which side a flag sits on, the order of three different features going clockwise. If the option reverses it, it is a reflection and it is out. This costs about two seconds and it is free marks.
The rules behind this
Sourced to the official brochure rather than restated here, so there is one place to correct when the Council revises it.
Part B examines six named areas: Visual Reasoning, Logical Derivation, General Knowledge/Architecture and Design, Language Interpretation, Design Sensitivity and Thinking, and Numerical Ability.
Visual Reasoning — understanding and reconstructing 2D and 3D composition. Logical Derivation — decoding a situation or context and drawing conclusions. General Knowledge, Architecture and Design — current issues, important buildings, historical progression, innovation in materials and construction. Language Interpretation — meaning of words and sentences, English grammar. Design Sensitivity and Thinking — observing and analysing people, space, product, environment; semantics, metaphor, problem identification. Numerical Ability — basic mathematics and its association with creative thinking; unfolding space using geometry.
Source · verified 2026-07-16
Part B allows 108 seconds per question, presented one after another, on an adaptive engine.
90 minutes across 50 questions. The adaptive structure dates to NATA 2025 per the President's foreword in V2.0, which states that NATA 2026 continues it.
Source · verified 2026-07-16
What almost everyone believes
“You either have spatial ability or you do not — it is innate, so there is no point drilling it.”
Mental rotation is among the most reliably trainable skills in cognitive psychology. It improves with practice and the improvement persists.
This belief is doubly costly: it stops candidates drilling the one Part B area that most dependably responds, and it hands them a ready-made excuse when they find it hard. COA's line about a sensibility that cannot be instilled gets stretched to cover this, but it does not: the Council is warning against cramming facts, not claiming that spatial skill is fixed at birth. Someone who does ten minutes a day for a month will visibly outperform their own baseline.
Depending on how long you have
Foundation
Understand the skill. Months out, or starting from zero.
Drill this now, because it improves and the improvement holds. Ten minutes a day of rotation puzzles for a month will move you further than any amount of reading about technique. And build the physical intuition too: pick up objects, turn them, predict what you will see before you look. The skill is the same one Part A's 3D composition is testing, which makes this the best-value drill on the site.
Drill
The practice protocol. What to repeat, how often, how to score it.
Practise the viewpoint shift deliberately: never say rotate the object, say walk around it. Then track one feature. If you find yourself trying to hold the whole shape, you have reverted — stop and restart the question the other way. And check chirality first on every question; the mirrored distractor is free elimination.
Exam-Day
What to actually do under the constraint — 108 seconds, no instruments, one pass.
Chirality first — scan for a mirrored option and delete it before doing any work. Then find the one distinctive feature and track only that. Do not spin the whole object; move yourself around it. If you are past sixty seconds still holding the shape in your head, take your best elimination and commit.
Try it
Fifteen minutes, and you will need a physical object — a book, a mug, anything with a clear front and an asymmetric feature.
- 01Hold the object. Look at it. Now close your eyes and predict what the left side looks like.
- 02Open your eyes and check by walking round it — not by turning it. Notice which is easier.
- 03Now do the reverse: keep your head still and rotate the object 90 degrees. Predict, then check.
- 04Most people find walking round dramatically easier. That asymmetry is the whole technique.
- 05Finish on the mock: run the Visual Reasoning questions and, before working any of them, scan for a mirrored option and eliminate it on sight.
The short version
Visual Reasoning is the best return on drill time in Part B, because mental rotation genuinely trains and the gains hold. Move your viewpoint rather than spinning the object — you have a lifetime of walking round things and none of spinning them. Track one distinctive feature instead of the whole shape. And know that rotation never changes handedness, so any mirrored option can be deleted on sight, in about two seconds, without visualising anything.
Next: mirrors and handedness properly — the property rotation cannot touch.
Questions people actually ask
- Can spatial ability be improved, or is it innate?
- It improves. Mental rotation is one of the most studied and most reliably trainable spatial skills, and practice gains persist. The belief that it is fixed is costly twice over: it stops candidates drilling the Part B area that responds best, and it supplies an excuse when the questions are hard.
- What is the fastest way to answer a rotation question?
- Check chirality first: rotation never changes handedness, so any option that is a mirror image can be eliminated on sight without visualising anything. Then track one distinctive feature through the rotation rather than the whole shape, and move your viewpoint rather than spinning the object.
- Why is it easier to imagine walking around an object than rotating it?
- Because you have spent your life doing one and not the other. Rotating an object in your head requires holding the whole shape in working memory and transforming it simultaneously; shifting your viewpoint leaves the object fixed and reuses spatial intuition you already have. The geometry is identical and the cognitive load is not.