Inference questions are where careful readers separate themselves from fast readers. In the rest of Reading Comprehension, you can often find a line that states the answer almost verbatim. Not here. An inference question deliberately asks about something the author never quite said out loud — and then dares you to fill the gap yourself. The trap is that the LSAT wants you to fill it exactly, with nothing extra, when your instinct is to reason like a normal human and add a little common sense. Beat that instinct and these become some of the most reliable points on the section.

What Inference Questions Test

Inference questions test whether you can distinguish between what a passage states, what it logically requires, and what it merely suggests to a reasonable person. On the LSAT, only the first two get you the point. The correct answer to an inference question is something that must be true — or is so strongly supported it may as well be — if we take everything in the passage as given.

Think of it this way: if the passage is a set of premises, the correct answer is a conclusion you can draw without adding any new premise of your own. Every wrong answer requires you to smuggle in a fact the passage never provided. Your entire job is to notice when that smuggling is happening.

This is a different muscle than the 'main point' or 'author's attitude' questions. Those ask you to summarize. Inference asks you to deduce. The best readers slow down and ask, for each choice: 'Where in the passage does this come from, and does the text actually guarantee it?'

How to Recognize Them

Inference stems are among the most common in RC, and they announce themselves with soft, indirect language. Watch for:

• 'It can be inferred from the passage that...'
• 'The passage suggests / implies that...'
• 'The author would be most likely to agree with which one of the following?'
• 'Which one of the following is most strongly supported by the passage?'
• 'The passage provides the most support for which one of the following statements?'

Contrast these with detail-question stems ('According to the passage...' or 'The passage states that...'), which point to something explicitly written. If the stem uses hedged, indirect verbs — suggests, implies, indicates, most likely — you are in inference territory, and you should expect the answer to be a logical step away from any single sentence, not a paraphrase of one.

The Method

Use a consistent four-step attack. It feels slow at first and becomes automatic with reps.

1. Read the stem carefully and note the scope. Some inference questions are global ('the author would most likely agree...'), and some are local ('the passage suggests that the second study...'). A local stem tells you exactly where to go — return to that part of the passage before looking at answers.
2. Anchor in the text. If the question is local, reread the relevant sentences. If it's global, recall the author's main claims and stated positions. Do not pre-phrase a specific answer here — inference answers are too unpredictable to guess in advance. Instead, arm yourself with the passage's actual claims.
3. Test each choice against the text. For every answer, ask: 'Does the passage guarantee this?' Not 'could this be true?' — almost anything could be true. Your standard is support, not possibility. Eliminate any choice that requires a fact the passage didn't give you.
4. When two survive, pick the weaker one. A choice that says 'some,' 'may,' 'at least one,' or 'is one factor' is easier to support than a choice that says 'all,' 'must,' 'the primary cause.' Cautious claims win inference questions; bold ones need more support than a passage usually provides.

The discipline in step 3 is everything. Read each choice as a suspect. Make it prove it belongs in the passage.

The Common Traps

The LSAT builds wrong answers to inference questions from a small, repeatable menu. Learn to name them and you'll spot them faster:

• Too strong / goes too far. The passage supports a modest claim, and the answer overstates it. The text says a policy 'contributed to' an outcome; the answer says it 'was the primary cause.' Watch every absolute word: all, never, only, must, cannot, always, the most.
• Outside knowledge. The answer is true in the real world, and your brain nods along — but the passage never established it. Correct-in-life is not correct-on-the-LSAT unless the passage put it there.
• Reversed or distorted relationship. The passage says X caused Y; the answer says Y caused X, or that X and Y are unrelated. Direction, order, and comparison are where distortions hide.
• Right idea, wrong subject. The answer describes something true of Group A but attaches it to Group B, or applies a claim from one paragraph to the topic of another. On passages that compare two views or two studies, this trap is everywhere.
• Half-right. The first half of the choice is perfectly supported, and the second half sneaks in an unsupported claim. Read the entire choice — a single unsupported clause kills the answer.
• Plausible but unsupported. The gentlest trap: nothing wrong with it, it just isn't backed by the text. This is why 'could be true' is the wrong standard — many wrong answers could be true.

Worked Example

The following passage and question are original Lovare practice, written in the style of the LSAT.

Early studies of urban tree canopy focused almost exclusively on aesthetic value, treating shade trees as ornamental features of city design. More recent research, however, has quantified the measurable effects of dense canopy on local temperature: streets shaded by mature trees can run several degrees cooler than adjacent unshaded streets during summer heat. This finding has prompted some planners to reclassify canopy as public infrastructure rather than decoration. Critics of the reclassification do not dispute the cooling data; they argue instead that infrastructure designations trigger maintenance obligations that many municipalities are financially unprepared to meet.

Question: The passage suggests that the critics of reclassification would most likely agree with which one of the following statements?

Choices:

• (A) Urban trees provide no measurable cooling benefit to city streets.
• (B) Reclassifying tree canopy as infrastructure could impose costs that some municipalities cannot currently afford.
• (C) Aesthetic considerations should remain the primary basis for urban tree policy.
• (D) Municipalities that reclassify canopy will inevitably fail to maintain it.
• (E) The cooling effect of shade trees has been overstated by recent research.

Answer: (B). The passage explicitly tells us the critics 'do not dispute the cooling data' and that they worry about maintenance obligations 'many municipalities are financially unprepared to meet.' (B) restates exactly that worry in cautious language ('could impose costs that some municipalities cannot currently afford') — fully supported.

Why the others fail: (A) and (E) contradict the passage — the critics accept the cooling data, so they don't deny the benefit or call it overstated. That's the 'too strong / contradiction' trap. (C) imports a position (aesthetics should be primary) the critics never take; the passage says nothing about their view on aesthetics — outside the four corners. (D) is 'too strong': 'inevitably fail to maintain it' is a much bolder claim than 'unprepared to meet' the obligation. Note how (B) is the weakest, most cautious answer, and it wins. That's the pattern.

A Few to Try

Work these the same way: anchor in the text, test each choice for guaranteed support, and favor the cautious claim. Full explanations are provided. These are original Lovare practice questions.

Read the following short passage and answer the two questions that follow:

Linguists once assumed that a child's rate of vocabulary growth depended chiefly on the sheer quantity of words the child heard. A series of longitudinal studies has complicated this view. In these studies, children who heard fewer total words but more frequent direct conversational turns — moments in which an adult responded to the child's own vocalizations — developed vocabulary at least as quickly as children exposed to a greater raw volume of speech. The researchers stopped short of claiming that raw volume is irrelevant, noting only that responsiveness appears to be a stronger predictor than they had expected.

How to Drill This Type

Inference is a skill built through volume plus honest review — not through passive rereading. Here's how to actually get better:

• Do full passages, not isolated questions. Inference answers depend on the whole passage's structure, so drill in context. Time yourself at roughly 8–9 minutes per passage once you're past the learning phase.
• Keep a trap log. Every inference question you miss, name the trap: too strong, outside knowledge, reversed relationship, wrong subject, half-right, or plausible-but-unsupported. Within a few weeks you'll see your personal weakness — most people over-trust 'plausible' answers.
• Practice the 'prove it' habit out loud. For your right answer, say which sentence(s) support it. If you can't point to text, you got it right by luck, and luck runs out on test day.
• Use official, current-format material. The most valuable practice is real, released LSAT questions. Access them through LSAC's official platform, LawHub, which is the authoritative source for genuine PrepTests. Remember the current test is two scored Logical Reasoning sections and one scored Reading Comprehension section (plus one unscored section), scored 120–180. Logic Games was removed in August 2024, so RC now carries a larger share of your score — inference reps are well spent.

Do the reps, name the traps, and hold every answer to the 'must be supported' standard. Inference questions reward the reader who refuses to leave the passage.