What the UCAT Abstract Reasoning section actually measures
The UCAT Abstract Reasoning subtest presents candidates with collections of abstract figures and asks them to identify underlying patterns, relationships, and organisational rules. Unlike the Quantitative Reasoning or Verbal Reasoning sections, Abstract Reasoning does not draw on curriculum knowledge; instead, it assesses a candidate's ability to generate hypotheses, test them rapidly, and apply consistent logical rules under significant time pressure.
For medical and dental school admissions, Abstract Reasoning serves a specific function: it provides an indicator of clinical reasoning potential that is independent of prior academic achievement. Two candidates with identical academic transcripts may present very different Abstract Reasoning profiles, and admissions tutors use this information to distinguish between applicants whose cognitive profile aligns more closely with the demands of clinical practice.
Section format and the time pressure that defines Abstract Reasoning
The Abstract Reasoning section contains 50 questions delivered in a 12-minute window, providing approximately 14.4 seconds per question. This pacing places Abstract Reasoning among the most time-pressured subtests within the UCAT, and it is a section where strategic efficiency directly determines the score a candidate ultimately achieves.
The section is divided into four distinct question formats, each requiring a slightly different cognitive approach. Understanding these formats individually is the first step towards developing a reliable approach to each one. The four formats are Set Identification, Complete the Series, Fill in the Grid, and Analogies.
- Set Identification — presented with a target set of figures, candidates select which of four answer options belongs to that set.
- Complete the Series — shown a sequence of figures, candidates identify the pattern and select the next figure in the sequence from four options.
- Fill in the Grid — a 3-by-3 matrix with one figure missing; candidates select the figure that completes the logical pattern.
- Analogy questions — two figures share a relationship; candidates identify a fourth figure that shares the same relationship with a third figure.
Each format appears roughly equally distributed across the 50 questions, though the exact mix varies between test forms. A sound preparation strategy ensures a candidate is equally comfortable approaching any of these four question types.
Set Identification: pattern matching against a defined target
Set Identification questions present a target set, typically containing between two and four figures, alongside a contrasting set marked as the complement. Four answer options are then presented, and the candidate must determine which of those four belongs in the target set rather than the complement.
The primary challenge in Set Identification is identifying the defining rule that separates the target set from the complement. This rule can involve several dimensions simultaneously: the shapes present, their arrangement, their orientation, the presence or absence of particular colours, the number of elements, or the symmetry of the configuration.
A systematic approach proves most effective. First, the candidate should identify the most immediately apparent distinguishing feature — perhaps the figures in the target set all contain a particular colour while the complement does not. Having formed a hypothesis, the candidate then tests each answer option against this feature, eliminating those that do not fit. If the primary hypothesis proves insufficient — meaning two or more options remain — a secondary distinguishing feature must be identified and the elimination process repeated.
One useful heuristic is to focus on what is present in the target set but absent in the complement, rather than the reverse. This directional focus tends to reduce the number of simultaneous conditions a candidate is tracking.
When working through Set Identification questions, the most common error is attempting to identify every difference between the target set and the complement before answering. In most cases, one or two distinguishing features are sufficient to isolate the correct answer; excessive analysis wastes the limited time available.
Complete the Series: tracking transformations across a sequence
In Series Completion questions, candidates observe a sequence of figures and must identify the pattern that governs the sequence's progression. The underlying logic is almost always one of transformation: one or more attributes of the figures change in a predictable way as the sequence advances.
The candidate must identify not just the type of transformation but also its direction and magnitude. Common transformation categories include rotation (clockwise or anticlockwise by a fixed angle), translation (elements shifting position within the figure), colour change (elements gaining or losing a colour attribute), size alteration (figures scaling up or down), and pattern repetition (a cycle of two or more figures recurring).
Series questions frequently involve more than one simultaneous transformation, which increases the difficulty. A figure might rotate and change colour at the same time, or a shape might shift position while simultaneously changing its size. The candidate's task is to track each stream of transformation independently and then combine the results to predict the next figure in the sequence.
One critical distinction in Series questions is between continuous and discrete transformation patterns. In a continuous pattern, the transformation is consistent at every step — a figure that rotates by 45 degrees each time will show that rotation at every position in the sequence. In a discrete pattern, the transformation only becomes apparent at intervals — a figure that changes colour every second position in the sequence. Recognising whether the observed pattern is continuous or discrete prevents candidates from selecting an answer based on an incomplete reading of the sequence.
Fill in the Grid: the two-axis logic of the 3-by-3 matrix
The 3-by-3 grid question is widely regarded as the most cognitively demanding format within Abstract Reasoning. Nine positions form a square grid, with one position left blank. The patterns governing the grid operate on two independent axes simultaneously: across each row and down each column.
The defining principle of the matrix is that each row and each column follows its own consistent rule. A change that occurs across a row — for example, a shape rotating by 90 degrees and changing from blue to red — will be reflected identically in the missing position for that row, assuming all other conditions are met. Similarly, down any column, the same transformation logic applies independently of the row logic.
Because two independent axes are in play, candidates cannot simply observe a single row and assume the missing position follows that row's pattern. The column logic must also be satisfied, and in some cases these two logical streams point towards the same answer — but in others, they create apparent contradictions that require careful reconciliation.
When approaching a 3-by-3 grid, the most effective strategy is to focus on the row or column that contains the greatest amount of visible information. If two positions in a row are already filled, the candidate can often identify the row's transformation rule with reasonable confidence. The same process is then applied to the corresponding column. Where the row and column predictions coincide, the answer is clear. Where they diverge, the candidate must identify which axis provides the more reliable signal, typically by checking additional reference points elsewhere in the grid.
Analogy questions: identifying and transferring a relationship
Analogy questions present a pair of figures with an established relationship, then introduce a third figure alongside four answer options. The candidate must determine which answer option shares the same relationship with the third figure that the first two figures share with each other.
The difficulty in Analogy questions lies in the abstract nature of the relationship being tested. The relationship is rarely as straightforward as one figure being a rotated version of another. More commonly, the relationship involves a structural transformation — the way elements are positioned relative to one another, the relationship between filled and unfilled spaces, or the way a pattern on one side of a figure corresponds to a pattern on the other side.
To approach Analogy questions reliably, the candidate should first identify what is constant across the two figures — the structural skeleton that does not change — and then identify what varies — the transformation that connects the first figure to the second. Once both the constant and the variable are understood, the candidate examines the third figure, applies the same constant structure to it, and then applies the same transformation to determine the expected answer.
For example, if the first two figures show a small shape moving from the top-left to the bottom-right position while maintaining a fixed orientation, then the third figure's answer will involve the same positional shift for its own shape, regardless of the figure's initial orientation or colour.