CAT RC Question Types: Inference vs Tone vs Main Idea

VARC

You finish a passage on monetary policy feeling like you understood it. Then you get three of the four questions wrong. You reread, and the correct options look obvious in hindsight, which is the worst possible feedback because it tells you nothing about what to change. So you conclude your comprehension is weak and go read more editorials.

The diagnosis is usually wrong. Aspirants who miss CAT VARC questions after understanding a passage are rarely failing at reading. They are answering an inference question with main idea logic, or a tone question with detail logic. Each type has a different standard for what counts as correct, and using the wrong standard produces confident wrong answers.

Why RC Accuracy Stalls at the Question, Not the Passage

Comprehension and selection are different skills. Comprehension is what happens while you read: you follow the argument, notice the turns, and hold the author's position. Selection is what happens after: you take four statements written by a test setter and decide which one the passage licences. CAT scores only the second.

That distinction explains a pattern most aspirants recognise. Your accuracy on easy passages with hard questions is worse than on hard passages with easy questions, even though the hard passage felt worse to read. The difficulty CAT controls most tightly sits in the options, not the prose.

• You understood the passage and still picked an option the passage never claimed.
• The correct answer looks obvious afterwards, which means the reading was fine.
• Your errors cluster in one or two question types rather than spreading evenly.
• You spend most of your time on the last two options rather than on the passage.

Any two of those together point at selection rather than comprehension. It is worth checking before you commit another month to reading practice that will not move the number you care about.

Selection improves when you know what standard is being applied. An inference option can be true and still wrong if the passage does not support it. A main idea option can be accurate and still wrong if it covers only one paragraph. Naming the type is how you load the right standard before the options start pulling at you.

The Five Question Types CAT Actually Asks

Almost every question in a CAT passage falls into one of five buckets. The table gives the type, the tell in the stem, and the standard that decides the answer. Copy it out once, because the third column is what you will actually use.

Type	Stem tell	What makes an option correct
Main idea	Best summarises, primary purpose, central argument	Covers the whole passage, not one strong paragraph
Inference	Can be inferred, the author would agree, implies	Follows from the text with the smallest possible step
Tone or attitude	Attitude towards, tone of the passage, the author regards	Matches the author's qualifiers and adjectives
Detail or fact	According to the passage, the author states	Is traceable to a specific line, paraphrased
Structure or function	Serves to, the role of paragraph three, in order to	Describes what the sentence does, not what it says

Why the Third Column Does the Work

Two options often survive elimination because both are consistent with the passage. Consistency is not the test. Once you know that inference rewards the shortest step and main idea rewards the widest supported claim, the survivors stop looking equal and one of them starts looking out of scope.

The Stem Signal Test: Naming a Question in Five Seconds

The test is deliberately mechanical. It runs on the stem alone, before the options are read, because reading options first is what lets a plausible distractor set your standard for you.

• A stem with no verb of judgement and a specific noun is usually a detail question in disguise.
• Any stem naming a paragraph number is a structure question until proved otherwise.
• Stems that quote a phrase from the passage are asking about function, not meaning.
• Except and least questions invert the standard, so name the type first and flip afterwards.

Inference Questions: The Evidence Distance Rule

An inference is a step, and CAT rewards short steps. If option A needs one line of the passage and option B needs two lines plus a reasonable assumption, A is the answer even when B feels more insightful. The exam is measuring restraint, not intelligence.

This is why strong readers sometimes score worse than cautious ones here. A reader who knows the subject fills gaps automatically, and a filled gap is an assumption the passage never made. Under CAT's standard, that is a wrong answer produced by knowing too much.

The practical guard is a sentence test. For every surviving option, name the line or lines that force it. If you cannot point at text, the option is not an inference, it is a belief you brought with you. Aspirants who apply that test consistently usually gain two to three marks a section within a month, purely by declining options they previously found convincing.

Three Distractor Shapes in Inference Questions

• The overreach. True in the world, unsupported in the passage. Usually the most attractive option.
• The extreme. Right direction, wrong strength. Watch for always, never, only and must.
• The reversal. Uses the passage's own words with the relationship flipped. Slow readers catch these, fast readers do not.

Working through para summary questions sharpens the same restraint, because a summary that adds anything is wrong for the same reason an overreaching inference is. Aspirants who train the two together improve faster than those who treat them as separate chapters.

Tone Questions: Read the Adjectives, Not the Argument

Tone is carried by word choice, not by position. An author can argue firmly against a policy in measured language, and the tone is then critical but not hostile. Aspirants who answer from the argument pick the strongest negative option and lose the mark to a milder one.

The practical move is to underline the author's own evaluative words while you read: unfortunately, remarkably, merely, so-called, admittedly. Four or five of those across a passage give you the tone with no interpretation at all, and they take no extra time to notice.

The Evaluative Words That Carry Tone

• Concession markers. Admittedly, granted, to be fair. These signal a balanced rather than hostile author.
• Dismissal markers. Merely, so-called, supposedly. These push the tone towards sceptical or critical.
• Surprise markers. Remarkably, strikingly, curiously. These usually indicate interest rather than judgement.
• Regret markers. Unfortunately, sadly, alas. These indicate sympathy with the subject, not opposition.
• Hedges. Perhaps, arguably, tends to. A passage full of hedges is almost never described by a strong tone word.

Count them as you read rather than judging at the end. Five hedges and one dismissal is a cautious author, whatever the argument concludes, and that count settles most tone questions before you reach the options.

Main Idea Questions: Scope Is the Only Test

A main idea option fails for one of two reasons: it is too narrow, describing one paragraph well, or too broad, describing a subject the passage merely touched. Accuracy is rarely the issue. Coverage is.

The fastest check is a coverage sweep. Take the candidate option and ask which paragraphs it accounts for. If paragraph four is doing work the option ignores, the option is too narrow, however elegant it sounds. This takes about fifteen seconds and settles most deadlocks on its own.

Four Ways a Main Idea Option Fails

• Paragraph capture. It describes the most vivid paragraph and ignores the rest.
• Topic drift. It names the subject correctly but not what the author says about it.
• Borrowed conclusion. It states the final paragraph's proposal as though it were the whole purpose.
• Over-generalisation. It is broad enough to describe three different passages, which means it describes none of them.

Two Options Left: How the Type Decides the Tiebreak

The deadlock is where marks are won, and the type gives you a rule for each case rather than a feeling. Run the matching line below instead of rereading the passage a third time.

1. Inference deadlock. Pick the option that needs less text. Distance from the passage is the tiebreak.
2. Main idea deadlock. Pick the option that covers more paragraphs. Coverage is the tiebreak.
3. Tone deadlock. Pick the milder option unless the passage contains genuinely strong language.
4. Detail deadlock. Return to the line. One of the two options changes a qualifier.
5. Structure deadlock. Ask what the sentence does for the argument, not what it claims.

Timed practice is what makes these rules survive contact with a clock. A set of VARC practice questions done with the type named aloud before each attempt is a different exercise from the same set done normally, and the accuracy difference usually shows within two weeks.

Build the Habit Into Your CAT Preparation Week

Two weeks of deliberate naming is usually enough for the habit to become automatic. Keep the volume low and the labelling strict, because the point is the label rather than the number of passages.

1. Week one. One passage a day. Label every question before reading options, and log the label next to your answer.
2. Week two. Same volume, but write the standard you applied for every question you got wrong.
3. Ongoing. Review the log weekly and count how many errors were type errors rather than reading errors.

Once the labels are stable, move to real papers, where stems are worded less helpfully than in practice sets. Topic-wise CAT exam previous year questions are the honest test of whether the habit holds. If your week has no fixed VARC slot at all, the AI study planner for CAT 2026 will place one and rebuild it when you miss a day.