Reasoning — Non-Verbal, Puzzles, and Data Interpretation

Why This Chapter Matters

Where the previous chapter dealt with reasoning built on words and statements, this chapter covers the other half of the reasoning syllabus: questions built on figures, spatial arrangements, structured puzzles, and tabulated data. These question types consistently separate well-prepared candidates from the rest, because they cannot be solved through memorised formulas alone — they demand a calm, methodical approach under time pressure, which is exactly the skill this chapter is designed to build. Puzzles and data interpretation questions in particular tend to appear in sets of four or five questions built around a single arrangement or table, meaning that cracking the underlying setup correctly unlocks several marks at once, while a wrong initial setup costs you the entire set — making the setup-building technique taught here one of the highest-leverage skills in your entire preparation.

Non-Verbal Reasoning: Series, Analogy, and Classification with Figures

Non-verbal reasoning questions replace words and numbers with geometric figures, and test the same underlying skills — pattern extension, relationship identification, and odd-one-out detection — using shapes, lines, shading, and rotation instead. Figure series questions present a sequence of figures that change according to a consistent rule across each step, and ask you to identify the next figure in the sequence. The most common transformation rules include rotation of the whole figure or of an internal element by a fixed angle each step, progressive addition or removal of a component (an extra line, dot, or shape appearing or disappearing in sequence), a consistent shift in shading or fill pattern, and a steady increase or decrease in the number of sides or internal elements. The effective technique is to isolate one attribute of the figure at a time — first track only how the outer shape changes, then separately track only how the shading changes, then separately track only how any internal marks change — rather than trying to absorb the whole figure's transformation at once, since most figure series combine two or three independent changes happening simultaneously, and untangling them one attribute at a time is far more reliable than holistic pattern-spotting.

Figure analogy questions present a pair of figures related by some transformation, and ask you to find a second pair related by the same transformation. As with verbal analogies, the discipline of naming the transformation explicitly — "the second figure is the first rotated ninety degrees clockwise with the shading inverted" — before scanning the options prevents you from being misled by an answer choice that merely looks superficially similar to the first pair without sharing the actual rule.

Figure classification, or "odd figure out," questions ask you to find which one figure among a group does not share a property common to the rest. Since figures can be classified by many different attributes simultaneously — number of sides, number of internal lines, symmetry, whether the figure is open or closed, direction of shading — the reliable approach is to check the group against several candidate attributes in turn: first count sides across all figures, then count internal elements, then check symmetry, and continue until you find the attribute by which exactly one figure breaks the pattern shared by the rest.

Mirror-image and water-image questions ask you to identify how a given figure or text would appear when reflected in a vertical mirror (mirror image) or reflected as if in still water below it (water image). The essential rule to internalise is that a mirror image reflects left-right (a mirror placed vertically along the right edge of the figure swaps the figure's left and right, leaving top and bottom unchanged), while a water image reflects top-bottom (as if the figure were flipped upside down and placed below itself, leaving left and right unchanged). A frequent error is confusing which axis inverts in each case, so it is worth fixing firmly in memory that "mirror flips sideways, water flips upside down" as a simple verbal anchor. When the figure includes letters or numbers, remember additionally that certain characters look identical or nearly identical to their mirror or water image (such as the letters A, H, I, M, O, T, U, V, W, X, and Y for vertical mirror symmetry), and recognising these can quickly eliminate wrong options.

Paper folding and cutting questions ask you to visualise a piece of paper being folded one or more times and then punched or cut, and to determine the resulting pattern of holes or the shape when the paper is unfolded. The reliable technique is to work backward one fold at a time from the final folded, punched state: unfold the last fold first, reflecting the punch marks across the most recent fold line to double them symmetrically, then unfold the previous fold and reflect again, continuing until all folds have been reversed. Attempting to visualise the entire unfolded result in one mental leap is error-prone; reversing fold by fold, reflecting the marks at each stage, is slower per step but dramatically more accurate, and with practice becomes fast enough for exam conditions.

Embedded figures and figure completion questions ask you to identify a smaller figure hidden within a more complex one, or to select the piece that correctly completes a larger figure with a section missing. These reward patient, systematic scanning — checking each answer option against the specific missing region's boundary lines and any pattern continuity (such as a line that must continue at the same angle, or shading that must match) — rather than a quick visual guess, since the distractor options are deliberately designed to be very close approximations of the correct piece.

Seating Arrangement Puzzles: Linear and Circular

Seating arrangement puzzles describe a group of people seated in a row (linear arrangement) or around a table (circular arrangement), give a series of clues about their relative positions, and ask you to determine the complete seating order or answer specific position questions. These puzzles are, in effect, a controlled logic exercise where you must place every person into a unique slot consistent with every given clue simultaneously, and the single most important habit for solving them efficiently is to read through all the clues once before drawing anything, mentally sorting them into "definite" clues that fix a position absolutely (such as "D sits at the extreme left end") and "relative" clues that only fix a relationship between two people (such as "B sits immediately to the right of C"). Placing the definite clues first, then working through relative clues that connect to already-placed people, and only at the end handling the vaguer clues (such as "E does not sit next to F") as elimination filters, produces a much faster and more accurate solution than trying to place people in the order the clues happen to be listed.

For circular arrangements, an essential early decision is whether people are facing the centre of the table or facing outward and away from the centre, because this single detail flips the meaning of "left" and "right" for every person in the puzzle — when facing the centre, a person's right hand points in the clockwise direction around the table, whereas when facing outward, a person's right hand points in the counter-clockwise direction. Missing this detail, or the question failing to specify it and requiring you to infer it from context, is the most common source of otherwise-correct-logic errors in circular arrangement puzzles, so it is worth explicitly noting the facing direction at the very top of your rough work before placing a single person.

Linear arrangement puzzles occasionally use two parallel rows facing each other (for instance, one row facing north and a second row facing south), and here the same left-right flip issue arises: a person's left hand in a row facing north points to the west, while a person's left hand in the opposite row facing south points to the east. Always establish and label the facing direction and the corresponding left-right convention for each row before working through the clues.

A powerful general technique across all seating puzzles is to use a placeholder grid with all possible seat numbers laid out on your rough sheet before you begin, and to fill in confirmed placements while marking tentative or excluded placements lightly (for instance, a small cross to note "clue three rules out B in seat 4"), so that as clues accumulate you are visually narrowing down possibilities rather than repeatedly redrawing the whole arrangement from memory. When a clue only narrows a person's position to two or three possible seats rather than fixing it exactly, it is entirely normal and expected — hold that ambiguity lightly on paper and resolve it once later clues intersect with it, rather than guessing prematurely and having to backtrack.

Puzzle Variants: Floors, Boxes, Scheduling, and Comparisons

Beyond seating, the exam frequently uses the identical logical skeleton dressed up as different scenarios: people living on different floors of a building, boxes of different colours or sizes stacked or arranged in a row, people or events scheduled across different days of the week or months of the year, and people compared by attributes such as age, height, weight, marks, or salary. Floor-based puzzles work exactly like linear arrangement puzzles turned vertical — treat "above" and "below" the way you would treat "left" and "right" in a row, and watch carefully for whether floor numbering starts from the ground floor upward or is described in some other order, since a puzzle stating "the ground floor is numbered one, and floors are numbered upward" behaves very differently from one where the topmost floor is numbered one.

Box and stacking puzzles usually add an extra dimension by also assigning each box an attribute like colour, weight, or brand, essentially combining a linear or vertical arrangement puzzle with an attribute-matching puzzle. The clean way to handle these is a small grid or table with one column for position and separate columns for each additional attribute, filled in progressively exactly as with seating arrangements — resist the temptation to track colour and position in your head simultaneously, since combined puzzles are specifically designed to overload unaided memory.

Scheduling puzzles distributing people or events across days of the week require the same definite-clue-first approach, with the additional wrinkle that "before" and "after" in scheduling puzzles are directional exactly like left-right in linear arrangement, so treat the days of the week as a simple linear row (Monday through Sunday, or whatever range is specified) and apply identical logic.

Comparison puzzles, where you must rank people by age, height, marks, or a similar quantity based on a series of comparative statements ("A is older than B but younger than C," and so on), are best solved by building a single ordered list from tallest/oldest/highest to shortest/youngest/lowest, inserting each person into the list as their relative position becomes clear from a clue, and treating ties or ambiguous positions the same way as with seating puzzles — hold the ambiguity visibly and resolve it as later clues arrive. A frequent higher-difficulty variant states comparisons indirectly through arithmetic ("A's marks are double B's marks, and C scored 15 more than A"), which requires you to also track the actual numeric values, not just relative order — for these, set up simple variable relationships and solve them as you would a basic algebra problem before returning to the ranking.

Data Interpretation: Reading Tables, Graphs, and Charts Accurately

Data interpretation, commonly abbreviated DI, presents information in a table, bar graph, line graph, or pie chart and asks a series of questions requiring you to extract values, compare quantities, or perform calculations such as percentages, ratios, and averages based on that data. The most important — and most frequently underrated — skill in DI is not calculation speed but careful, accurate reading of the given data structure before attempting any question. Spend the first thirty to forty seconds after seeing a DI set simply understanding what the table's rows and columns represent, what units the graph's axes are measured in, and what the pie chart's total represents, because a misread axis label or unit produces a series of confidently wrong answers across every question in the set, and this initial-reading error is far more common and far more damaging in DI than any arithmetic mistake.

For tabular DI, watch specifically for whether values are given as absolutes or as percentages of some total, and whether a "total" row or column is provided directly or must be calculated by summing the visible entries — many DI sets deliberately omit the total to test whether you notice its absence and calculate it correctly rather than assuming a round number. For bar graphs, pay close attention to the y-axis scale, particularly whether it starts at zero or at some other baseline and what value each gridline interval represents, since bar graphs are a favourite place for the exam to test whether you read the scale correctly rather than estimating bar heights visually.

For line graphs, which typically track a quantity's change over time, the key reading skill is distinguishing between the value at a point (read directly off the y-axis) and the trend between two points (the slope, indicating rate of increase or decrease), since many questions specifically ask about the period of "maximum increase" or "maximum percentage change," which is not necessarily the period ending at the highest absolute value — a common trap is assuming the highest point on the graph corresponds to the greatest increase, when the greatest increase might actually occur during a steep climb toward a lower final value.

For pie charts, remember that the values shown are proportions of a whole circle representing 360 degrees or, more commonly in percentage-based charts, 100 percent, and that converting a given percentage to an actual quantity requires knowing or calculating the total value the pie chart represents — a total that is very often given separately in a line of text above or below the chart itself, easy to overlook if you move straight to the chart. When two or more pie charts are given together (for instance, showing the same category breakdown for two different years), questions frequently ask you to compare an item's actual value, not merely its percentage share, across the two charts, which requires you to compute two absolute values using each chart's own total before comparing them — comparing the raw percentages directly, without converting to actual values, is one of the most common DI errors when totals differ between the charts.

Calculation efficiency in DI comes from strong percentage, ratio, and approximation skills carried over from quantitative aptitude, but a distinctly reasoning-flavoured skill specific to DI is estimation before precise calculation: scanning the answer options first to see how far apart they are lets you judge whether a quick approximate calculation is sufficient (when options are widely spaced) or whether precise calculation is unavoidable (when options are closely clustered), and this single habit alone can save a meaningful amount of time across a full DI set without sacrificing accuracy.

Integrating Speed and Accuracy Across the Non-Verbal Syllabus

The topics in this chapter share a common thread: each one rewards a slow, careful setup phase followed by fast, mechanical execution, rather than rewarding either pure speed or pure caution alone. Candidates who rush the setup phase — sketching an arrangement grid hastily, skimming a DI table's headers, or glancing at a figure series without isolating individual attributes — tend to solve quickly but inaccurately, while candidates who are careful throughout, including during the mechanical execution phase, tend to be accurate but too slow to complete the section within the allotted time. The winning approach, developed through repeated timed practice, is to front-load your carefulness into the setup phase specifically, and then trust that careful setup enough to move quickly and confidently through the remaining questions in that set.

Because puzzles and DI both typically appear as multi-question sets built on one shared setup, a further strategic point is worth internalising for the actual exam: if, after a reasonable effort of a minute or two, you cannot resolve a puzzle's setup completely, do not abandon it outright — check whether the specific question being asked can be answered from the partial information you have already worked out, since many puzzle sets include at least one or two questions answerable without the full arrangement being pinned down. Skipping the entire set purely because the last one or two positions remain ambiguous can mean leaving free, gettable marks on the table. Consistent, deliberate practice across figure reasoning, arrangement puzzles, and data interpretation, combined with the setup-first discipline emphasised throughout this chapter, will make this often-feared portion of the reasoning syllabus into one of your most reliable sources of marks on exam day.