Reading comprehension is the engine room of the LNAT Section A multiple-choice paper, and the section on which an applicant's overall LNAT score leans most heavily. Section A asks candidates to work through roughly twelve passages of argumentative, journalistic, and legal-academic prose, answering three to four multiple-choice questions on each, inside a single 95-minute window. That gives a candidate something close to two minutes and fifteen seconds per question, and the passages themselves are not short. The 42 questions are not testing reading speed in any simple sense. They are testing whether a candidate can extract the load-bearing claims of an unfamiliar argument, distinguish the author's view from the view being reported, and predict how a careful author would respond to a hypothetical objection. Everything that follows on this page is built around that definition.
The architecture of the LNAT Section A paper
Before any candidate can plan a sensible preparation strategy, they need an accurate mental model of what the paper actually contains. Section A is not a reading-speed test and it is not a vocabulary test. The passages are drawn from real journals, op-ed pages, legal philosophy collections, and policy reviews. They run anywhere from a tight three-paragraph editorial to a six- or seven-paragraph academic argument. The text type matters: a candidate who approaches a polemical newspaper column the same way they approach a careful philosophical reconstruction will lose minutes to confusion.
Each passage carries a small fixed set of questions, usually three or four, and those questions cluster around the argumentative spine of the passage. The first question on a passage often asks something literal — a paraphrase of a claim, a definition in the author's own terms. The middle question tends to be inferential, asking what follows from a particular sentence or which assumption is required. The last question tends to be evaluative, asking the candidate to weaken, strengthen, apply, or critique a specific move. Recognising that internal rhythm is a genuine exam skill, not a generic test-taking slogan.
The 95-minute budget is the second architectural fact a candidate must internalise. Twelve passages at roughly eight minutes each leaves a thin margin, and the first two passages almost always cost a candidate more time than the last two because of unfamiliar vocabulary and denser sentence structure early in the paper. For most candidates I work with, the single most productive shift in preparation is treating the first three passages as a warm-up where they are willing to spend up to ten minutes each, accepting that the back half of the paper will move faster because the argumentative moves begin to feel repetitive.
The four question archetypes that decide an LNAT Section A score
Almost every multiple-choice question in Section A falls into one of four families. Candidates who learn to name the family before they read the options move faster and miss fewer.
- Literal-paraphrase questions. These ask the candidate to restate, in the stem's own language, a claim the author has already made. The trap options are usually subtly broader, narrower, or more evaluative than the passage. The tactical move is to find the precise sentence in the passage and read it twice before looking at the options.
- Inference-and-assumption questions. These ask what must be true, what is most likely true, or what the author is committed to. The correct option is one the author would have to accept to keep their argument standing. The distractors are tempting because they are plausible opinions the author might privately hold, but plausibility is not commitment.
- Argument-structure questions. These ask which option best describes the function of a sentence, paragraph, or sub-argument. The candidate has to read the passage as a structure, not a sequence. The single most common error is selecting an option that is true about the passage's topic but does not describe the function of the named element.
- Evaluative-application questions. These ask the candidate to weaken, strengthen, or extend a particular claim. The stem names the claim precisely, and the candidate's job is to find the option that does the named job to that precise claim, not to a related but broader claim elsewhere in the passage.
For most candidates, argument-structure and evaluative-application questions are where the score separates. Literal-paraphrase questions are often answered correctly by readers across the ability range, because the passage has done most of the work. Inference and assumption questions reward careful reading, and the evaluative family rewards a kind of cold logical triage that not every strong reader has practised. Preparation that treats the four families as a single undifferentiated 'comprehension' task will leave the two harder families under-trained.
How to read the stem before the passage
A surprisingly powerful move is to read the question stems first, then the passage, then the stems again. The first read primes the candidate to mark up the passage for the specific features each stem cares about. The second read of the stems is when the candidate confirms or revises the answer. This is not a speed trick; it is a re-allocation of attention. The first-pass reading is converted from passive absorption into targeted extraction, which is what 95 minutes of MCQ work actually demands.
Argument mapping: the skill that lifts a Section A score by three to five marks
Argument mapping is the practice of representing a passage as a set of linked claims rather than a sequence of sentences. For a candidate who has never done this, the work looks odd at first. It is not annotation in the literary sense; it is diagrammatic. A good working method is to mark each paragraph with a one-line summary in the margin, then draw arrows between the summaries where the author signals 'because', 'therefore', 'however', or 'on this view'. After twenty minutes of practice on a single passage, the candidate has a map that is shorter than the passage and tells them where to look when a question names a specific move.
The reason argument mapping lifts a Section A score is that it gives the candidate a fast lookup table. A question that asks 'which of the following would most weaken the author's claim in paragraph three' becomes a question about a specific node on the map. The candidate does not have to re-read the passage; they read the relevant node and its incoming arrows. For most candidates, this shaves between 30 and 60 seconds off the harder questions, which compounds across 42 questions into a meaningful time reserve.
Argument mapping also exposes a category of error that pure re-reading cannot. Many wrong answers in Section A are wrong not because the candidate failed to understand the sentence in question but because the candidate answered a question about the wrong sentence. The map forces the candidate to confirm that the option they are about to select refers to the same claim the stem refers to, in the same paragraph, in the same argumentative role. That single check catches a high proportion of avoidable errors.
Working example: mapping a two-paragraph editorial
Take a short editorial arguing that public funding for the arts should be defended on civic rather than purely economic grounds. Paragraph one sets out the orthodox economic case, paragraph two argues that the civic case is independent of and superior to the economic case, paragraph three offers two objections and responds to each. A well-built map might label paragraph one 'orthodox economic case', paragraph two 'civic case as independent and superior', paragraph three 'two objections answered'. The arrows run from paragraph one to paragraph two with a 'but' marker, and from paragraph two to paragraph three with an 'even so' marker. A question asking 'which of the following would most strengthen the author's position in paragraph two' now points the candidate at a single labelled node, and the correct option will be one that supports the independence claim, not the civic value claim. A common wrong answer will strengthen the civic value claim, which the author has not asked the candidate to defend.
Opinion spotting versus opinion paraphrasing: a distinction that decides the inference family
The single most common error I see in Section A work is conflating two operations: spotting an opinion the author mentions, and paraphrasing the author's own opinion. The passages are full of opinions the author is reporting, attacking, comparing, or setting aside. A strong inference question will usually ask about the author's view, not the reported view, and the distractors will be drawn from the reported views.
The discipline that prevents the error is short and unglamorous. Before answering any inference question, the candidate should be able to point at a sentence in the passage and say, in their own words, what the author believes about the topic of the question. If they cannot do that, they are about to answer from the wrong voice. In my experience this usually takes a candidate an extra ten or fifteen seconds on the question, but it converts a coin-flip answer into a defensible one.
A related sub-skill is the recognition of 'scaffolding' sentences, which exist to hold up the author's argument rather than to express it. Sentences that begin 'To see this, consider…', 'There are two reasons for this', or 'A common objection goes as follows' are scaffolding, not claim. A question that targets a scaffolding sentence is usually a structure question, not a content question, and the candidate who treats it as content will pick an option that paraphrases the scaffolding example instead of describing its function.
Time budgeting across 12 passages and 42 questions
Time budgeting is where most LNAT Section A preparation falls down, and it is also where a small amount of disciplined practice pays off disproportionately. A workable budget for a 95-minute paper is roughly nine minutes per passage for the first three passages, eight minutes per passage for the middle six, and seven minutes per passage for the last three. That gives a candidate a buffer of three to four minutes for any single difficult passage, and it ensures the final passages are not abandoned in a last-minute scramble.
Within each passage, the budget should be allocated across the questions rather than across the reading. A common mistake is to spend five minutes reading a passage and one minute on the questions. For most candidates the inverted ratio produces a better score, because the questions are where marks are actually won or lost. A two-pass reading works well: a first read of two to two and a half minutes to mark the argumentative spine, a second targeted read of one minute after the stems have been read, then two minutes for the four questions.
Guessing strategy matters at the margin. There is no negative marking in Section A, which means an unanswered question is a guaranteed zero and a guessed question has the underlying base rate. For most candidates the base rate on a five-option LNAT question sits somewhere between twenty and twenty-five per cent, and it is rare for a candidate's educated guess to fall below that range. Leaving a question blank should be reserved for moments when the candidate has run out of time on a passage and cannot make a defensible elimination. Even then, a quick mark of the option that is most consistent with the passage's overall position is a stronger move than leaving the row empty.