CAT VARC Inference Questions: How to Tell Inference From Fact and Assumption

CAT VARC Inference Questions: How to Tell Inference From Fact and Assumption

Published: June 9, 2026 | 12 min read | CAT Preparation

Look at any honest VARC error log and one pattern keeps showing up. Inference questions in CAT VARC are the single most wrong-answered RC question type, and the reason is not weak reading. It is a quiet confusion about what the question is even asking. Aspirants treat "what can be inferred" as "what is stated," or they reach for the option that feels deepest, when the right answer is usually the modest one sitting half a step beyond the text. That gap between reading well and answering precisely is where easy marks leak out, paper after paper. This guide gives you a clear three-way test to close it.

Why Inference Questions Trip Up Strong Readers

Here is the part that catches people off guard. The aspirants who struggle most with inference are often the strongest readers in the room. They understand the passage fully, so they keep adding their own knowledge to it. That extra context is exactly what gets them the wrong option.

An inference question is not asking what you know. It is asking what the passage forces you to conclude. The moment you bring in a fact from outside the text, however true it is in the real world, you have left the boundary the question cares about. Strong readers cross that line because the passage feels like a starting point rather than a fence.

So the real fix is not reading harder. It is staying inside the passage and taking the shortest logical step it allows. To do that reliably, you need to know precisely what counts as a fact, what counts as an inference, and what counts as an assumption. That is the three-way test.

The 3-Way Distinction: Fact vs Inference vs Assumption

Almost every wrong answer on an inference question is one of these three things wearing the wrong label. Get the categories clear and the option set starts to sort itself. Here is each one, defined crisply, with a tiny example built on a single sentence.

Take this line as the source text: "The library extended its hours during exam week, and footfall doubled."

Notice the direction of each one. The stated fact just repeats the line. The inference points forward, a conclusion the evidence pushes you toward. The assumption sits underneath, a hidden support the author leans on without saying so. In an inference question, only the middle one is the answer. The other two are the most common decoys.

• Stated fact decoy. It is true and findable in the passage, so it feels safe, but inference stems do not want a quote. They want the next step.
• Assumption decoy. It is also unstated, which makes it easy to confuse with an inference, but it comes before the argument, not after it.
• The real inference. Unstated, forward-pointing, and forced by the lines. Modest, not dramatic.

This same forward-versus-underneath instinct shows up across RC. It is close cousin to reading the author's tone, where you also judge what is implied rather than stated, and to spotting main idea questions, where the trap is picking a true detail instead of the central point.

The Elimination Technique for Inference Options

Most aspirants try to pick the right option directly. On inference questions, eliminating is faster and far more reliable. Three filters knock out almost every wrong choice, and the option that survives all three is the one the passage actually forces.

• Too strong. The option uses absolute words like always, never, all, only, or proves, or it states a conclusion far bigger than the lines support. The passage rarely guarantees that much, so it goes.
• Too weak. The option is so vague or so safe that it would be true of almost any passage on the topic. If it tests nothing, it is not the inference the question wants.
• New information. The option adds a cause, a comparison, a number, or a detail the passage never mentions. It might be true in real life, but the text does not back it, so reject it.

The trick is to run these filters before you fall in love with any single option. Read all four choices, strike the too-strong and the too-weak, then check the rest for smuggled-in new information. Often two options survive the first pass and one of them sneaks in a detail the passage never gave you. That detail is your tiebreaker.

One more habit helps here. When two options look equally valid, prefer the weaker, more cautious one. Inference answers almost never overclaim, so between a bold conclusion and a careful one that both seem supported, the careful one is usually right. You can watch that accuracy climb in a tool that tracks accuracy by question type.

A Worked Example, Step by Step

Theory clicks once you walk a real stem. Here is a short passage extract and four options. Read it the way you would in the exam, then watch the elimination run.

Passage: "For decades, the town's economy ran on a single textile mill. When cheaper imports arrived in the 1990s, the mill cut shifts and then closed. The local council, which had long resisted new industry to protect the mill, now courts software firms with tax breaks."

Stem: It can be inferred from the passage that the council's attitude to new industry:

1. has reversed compared to its earlier stance.
2. was always hostile to all forms of business.
3. changed because software pays higher salaries than textiles.
4. was influenced by outside factors at some point.

Run the filters. Option 2 says "always" and "all forms of business," which is too strong; the council protected the mill, so it was not hostile to all business. Option 3 introduces new information, since the passage never mentions software salaries or compares pay. Option 4 is too weak, so vague it would fit almost any economic passage. Option 1 is what is left, and it is forced by the text: the council resisted new industry before and courts it now, so its stance has reversed. Modest, supported, correct.

Run that sequence enough times and it stops being a checklist. You start hearing "too strong" the instant you read an absolute word, and "where is that in the passage" the instant an option names a detail you do not remember reading.

How to Practise Inference Under Timing

Knowing the method is not the same as applying it with ninety seconds on the clock and three RCs still to go. Inference accuracy collapses under time pressure precisely because the elimination steps are the first thing aspirants drop when they panic. The fix is to build the steps into how you practise, not just how you review.

• Tag every inference miss. After each set, note whether you fell for too strong, too weak, or new information. Patterns appear fast, and most people have one repeat offender.
• Time the elimination, not the reading. Give yourself a fixed window per question and force all three filters inside it. Speed comes from the filters being automatic, not from rushing the passage.
• Do mixed sets. Practising inference questions back to back lets you cheat by pattern. Mix them with main idea and tone questions so you have to identify the type first, the way the real paper makes you.
• Review the wrong options, not only the right one. For each decoy, name which filter should have caught it. This is where the reflex is built.

Build this into your wider CAT preparation so inference stops being a coin flip and becomes a section where you gain marks. Calibrated practice matters more than volume here, because the difference between a CAT inference question and an ordinary one is how tight the right answer sits against the wrong ones. Keep your broader CAT 2026 preparation aimed at RC accuracy first, and you can always explore all CAT preparation blogs for the other RC question types as you go.

Optima Learn Editorial Team

Optima Learn is an AI-powered CAT preparation platform that builds personalised study plans, tracks your progress across topics, and adapts your roadmap as you improve.