Studio Matrx Monthly · Volume 1 · Issue 2 · July 2026
Amogh N P
 In loving memory of Amogh N P — Architect · Designer · Visionary 
NATA 2026 / Module 4Volume 1 · Issue 2 · July 2026
Part B · on screen4.4 · Visual Reasoning

Every drawing is a lie. Learn which one.

A plan tells you nothing about height. An elevation tells you nothing about depth. A section tells you everything about one slice and nothing about anywhere else. None of these is a failure — it is the deal. Each view destroys specific information in exchange for clarity about the rest, and Visual Reasoning is largely testing whether you know which information just went missing.

ByAmogh N P· Architect & interior designer7 min read · verified 2026-07-16
A carrot sliced clean through beside an apple cut in half on a wooden board, both exposed faces showing round cross-sections

Plan destroys height. Completely.

This is the one that catches people, and it caught us already in Part A: a composition whose logic is a plan does not read in the three photographs.

In plan, anything directly above anything else is invisible. Stack ten cubes and the plan shows one square. Put a tower next to a doorstep and, if their footprints match, the plan cannot tell them apart. Height information is not compressed in plan — it is gone.

So when a question gives you a plan and asks what the object could be, the honest answer is usually many things, and the question will be asking which of the options is consistent with it rather than which is determined by it. That distinction is the question. Candidates who look for the single right answer in a plan are looking for something the drawing does not contain.

The useful reflex: on seeing a plan, immediately ask what could be hiding under this? Every answer that shares the footprint is still alive.

A plan destroys height entirely, and the section shapes of common solids are recall rather than visualisation EVERY VIEW DESTROYS SPECIFIC INFORMATION. BY DESIGN. PLAN KILLS HEIGHT — COMPLETELY 4 stacked 1 cube SAME PLAN indistinguishable ASK: "WHAT COULD BE HIDING UNDER THIS?" EVERY OPTION SHARING THE FOOTPRINT IS STILL ALIVE. THE STEM USUALLY ASKS "POSSIBLE", NOT "CERTAIN". READ IT. SECTIONS ARE RECALL, NOT VISUALISATION CONE — the only solid giving all four conics horizontal → CIRCLE oblique thru side → ELLIPSE parallel to slant → PARABOLA vertical thru apex → TRIANGLE CYLINDER horiz → CIRCLE · vert → RECTANGLE · oblique → ELLIPSE SPHERE every cut → CIRCLE. always. nothing to work out. CUBE corner-to-corner across diagonals → REGULAR HEXAGON (surprises everyone — which is why setters like it) RECONSTRUCTION: ELIMINATE AGAINST EVERY VIEW. THE THIRD VIEW EXISTS TO KILL SOMETHING — FIND OUT WHAT.
Plan destroys height. Completely.

Sections have shapes, and the shapes are predictable

A section is what you see when you cut a solid and look at the exposed face. The useful part is that for simple solids, the section shape is entirely determined by the cut direction — so these are recall questions wearing a visualisation costume.

Cylinder. Horizontal cut: circle. Vertical cut through the axis: rectangle. Oblique cut through the curved wall only: ellipse. Those are the three, and the ellipse is the one people miss.

Cone. This is the classic, and it is worth knowing cold because it is the richest single solid in the area. Horizontal: circle. Oblique, cutting all the way through the side: ellipse. Parallel to the slant side: parabola. Vertical through the apex: triangle. These are the conic sections, and a cone is the only common solid that yields all of them.

Sphere. Every plane cut gives a circle, always. There is nothing to work out.

Cube. The interesting one. A cut can give a triangle, a rectangle, a pentagon, or — cutting corner to corner across the diagonals — a regular hexagon. The hexagon surprises everyone the first time and is a favourite of question-setters for exactly that reason.

Reconstructing a solid from views

The other direction: you are given two or three views and asked what the object is.

The method is elimination, not construction. Take each option and check it against every view. An option survives only if it is consistent with all of them — plan, elevation and side. Most options die on the second view, which is why this is faster than trying to build the object in your head from scratch.

And the crucial move, again: look for what is hidden. The single most common trap is an option that matches the plan and the elevation perfectly but has something concealed behind or beneath — a void, a notch, a step — that the third view exposes. If the question gives you three views, the third one exists to kill something. Find out what.

If you are only given two views, the object is usually not uniquely determined — and the question knows that. It will ask which is possible, not which is certain. Read the stem carefully, because possible and certain are different questions and the options are built to punish the confusion.

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.

OfficialNATA 2026 Information Brochure V2.0 · §4.0

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

What almost everyone believes

If I can picture the object clearly enough, I can read everything about it from the plan.

A plan destroys height information entirely. It is not compressed or implied — it is absent, and no amount of picturing recovers it.

Candidates treat views as windows onto an object rather than as projections that discard a dimension by design. Ten stacked cubes and one cube have the same plan; nothing in the drawing distinguishes them. So a question offering a plan usually asks which options are CONSISTENT with it, not which is determined — and a candidate hunting for the single right answer is looking for information the drawing does not contain. The productive reflex is to ask what could be hiding under this, and keep every option that shares the footprint.

Depending on how long you have

Foundation

Understand the skill. Months out, or starting from zero.

Cut things. Actual things — an apple, a carrot, a bar of soap. Cut a cone every way you can and watch the circle become an ellipse become a parabola. Ten minutes with a knife builds an intuition that no diagram does, and the conic sections stop being a list you memorised and become something you have seen.

Drill

The practice protocol. What to repeat, how often, how to score it.

Learn the section table cold — cylinder, cone, sphere, cube — because these are recall questions dressed as spatial ones, and recall is instant while visualisation is not. On reconstruction questions, practise elimination against every view rather than building the object, and always ask what the third view is there to kill.

Exam-Day

What to actually do under the constraint — 108 seconds, no instruments, one pass.

On a plan, ask immediately what could be hiding underneath. On a section question, recall rather than visualise — the cone gives all four conics and the cube can give a hexagon. On reconstruction, test each option against every view, and read whether the stem says possible or certain, because they are different questions.

Try it

Twenty minutes, and something you can cut. A carrot is ideal; soap is neater.

  1. 01Cut a cone-ish object horizontally: circle. Now at a shallow angle through the whole side: ellipse.
  2. 02Now cut parallel to the slant side: parabola. Now straight down through the apex: triangle. All four conics, in one vegetable.
  3. 03Take a cube of soap and cut it corner to corner across the diagonals. It gives a regular hexagon. This surprises everyone.
  4. 04Now build something from three or four blocks and draw its plan only. Hand it to someone and ask them to build it. They will get it wrong.
  5. 05Whatever they built that was not yours is exactly what the plan failed to say. That gap is what these questions test.

The short version

Every view destroys specific information by design: plan kills height completely, elevation kills depth, a section speaks only for its slice. Section shapes are recall, not visualisation — a cone gives all four conics, a sphere always gives a circle, and a cube can give a hexagon. On reconstruction, eliminate against every view rather than building the object, ask what the third view is there to kill, and read whether the stem says possible or certain.

Next: pattern completion — and how to find the rule before the clock finds you.

Questions people actually ask

What shape is the section of a cone?
It depends entirely on the cut, and a cone is the only common solid that gives all four conic sections. Horizontal gives a circle; an oblique cut through the whole side gives an ellipse; a cut parallel to the slant side gives a parabola; a vertical cut through the apex gives a triangle. These are recall, not visualisation — learn them cold.
Can a cube have a hexagonal section?
Yes — a plane cutting corner to corner across the diagonals produces a regular hexagon. It surprises almost everyone the first time, which is exactly why question-setters like it.
Why can I not tell what an object is from its plan?
Because a plan destroys height information entirely. Ten stacked cubes and a single cube share the same plan. So questions offering a plan usually ask which options are consistent with it rather than which is determined by it — and looking for a single right answer means looking for information the drawing does not contain.