GMAT Critical Reasoning is the part of the Verbal section where high scorers quietly separate themselves from the rest of the field. The questions look small, usually a 60 to 120 word argument followed by a stem and five answer choices, and the time pressure on the GMAT Focus makes it tempting to read them at speed. In my experience the candidates who break 84 in Verbal slow down precisely on these items, because the three question families that dominate the section, assumption, weaken, and strengthen, all ask the same underlying question in three different directions: where is the gap in the argument, and what would a new sentence do to that gap. Once you see the stimulus as a skeleton rather than a paragraph, the answer choices stop looking like five competing opinions and start looking like five testable moves on one structure.
The aim of this article is to give you a working map of that skeleton and to show how a single stimulus, read carefully once, can be re-used across assumption, weaken, and strengthen stems. We will work through stimulus anatomy, the three question types, the trap answer patterns, and a triage protocol you can use when a CR item refuses to give up its answer inside the 1 minute 30 second budget the GMAT Focus effectively allows on these passages.
The argument skeleton every GMAT Critical Reasoning stimulus hides
Every CR passage, regardless of topic, is built from four movable parts. There is a conclusion, the claim the author is trying to get you to accept. There is at least one piece of evidence, a fact, statistic, or observation that the author uses to support the conclusion. There is usually an unstated bridge, the logical link that says, in effect, this evidence is enough to carry this conclusion. And there is almost always some kind of context or counterweight, an alternative explanation, a competing cause, or a limit on scope that the author is ignoring or quietly acknowledging.
Your first job on any GMAT Critical Reasoning item is to identify which sentence is the conclusion and which is the evidence. The conclusion is not always the last sentence. In about a third of items, the conclusion sits in sentence one, and the evidence follows. In another third, a transitional phrase like therefore, thus, hence, or so flags the conclusion late. In the remaining items, the conclusion is reached by elimination: the only sentence that every other sentence seems to be feeding into is the conclusion.
Once conclusion and evidence are sorted, draw the bridge sentence in your head. The bridge is the step the author expects you to take for free. It is rarely written. A useful exercise is to finish the sentence and therefore after the evidence in your own words, then compare to the actual conclusion. If the logical gap between the two feels larger than a single inference, you have probably found the assumption. If the gap feels clean, the argument is unusually tight, and the question is more likely testing whether you can spot an alternative explanation or a scope problem.
Read the passage twice if you have to. The GMAT Focus CR passages are short enough that a second read is faster than chasing distractors you would have eliminated on a cleaner first pass. The candidates I work with who push past V84 almost always have a habit of re-reading the stimulus before they read the stem, not after.
Assumption questions: finding the unstated link the author needs
Assumption items ask, in the language of the test makers, what must be true for the argument to hold. Concretely, they ask: which unstated bridge is the author assuming between the evidence and the conclusion. The official phrasing on the GMAT is usually a variant of which of the following is an assumption that the argument depends on or which of the following must be true if the argument is to hold. The word must is doing all the work. If the choice is false, the argument falls. If the choice is true, the argument stands, or at least is no worse off.
There are three useful sub-types to keep separate. A causal assumption says that the evidence actually causes the conclusion, not merely correlates with it. A linkage assumption says that the evidence, drawn from one group or setting, applies to the group the conclusion is about. A negation assumption is one whose opposite would clearly destroy the argument. The negation test is the single most reliable tool in your CR kit. Take the candidate assumption, mentally invert it, and ask: does the argument still work? If yes, that assumption is not required. If no, you have a keeper.
Consider a representative shape. A city argues that a new bike lane caused a 12 percent rise in cycling commuters, based on counts taken at three intersections over six months. The conclusion is that the lane caused the rise. The evidence is the rise itself, plus a few intersection counts. The bridge the author needs is that no other factor, weather, fuel prices, a new employer, a parallel transit discount, could plausibly explain the same rise. The negation test exposes this: if weather was unusually clement during those six months, the rise is partly explained, and the causal claim weakens. So the assumption, in plain English, is that conditions other than the bike lane were roughly stable across the measurement window.
Trap answers on assumption items tend to be statements that are helpful to the argument but not necessary. They add extra support, fill in a side detail, or strengthen a different part of the passage. A common GMAT move is to write a choice that would be a great strengthen answer; do not pick it. The question is asking what the author must assume, not what would help the author if it were true. Another trap is the choice that is too broad, covering all possible cases, when the argument only needs a narrower bridge. The negation test catches both of these faster than any other method I have seen.
Weaken questions: attacking the bridge from the other side
Weaken items ask which choice, if true, would make the conclusion less likely to follow from the evidence. You do not need to destroy the argument entirely. A small dent in the bridge is enough. The official phrasing varies, but the signal words are which of the following, if true, would most weaken or would most seriously call into question. The word most matters: a partial weaken is correct only if no other choice is a stronger weaken.
The fastest way into a weaken item is to use the assumption skeleton you already drew. The correct weaken answer is, in most cases, a statement that directly attacks the same unstated bridge the author relies on. If the bridge is causal, the weaken answer is a plausible alternative cause. If the bridge is about a sample, the weaken answer is a fact about the sample that would skew the conclusion. If the bridge is a scope claim, the weaken answer is a counter-example from outside the claimed scope.
Watch for the classic weaken trap: a choice that attacks the evidence but not the bridge. The GMAT often includes a choice that questions whether the data is accurate, when the argument would still be vulnerable even with perfect data, because the bridge itself is the problem. Read for the bridge first, and the trap dissolves. Another trap is the choice that attacks an unrelated premise, a side observation the author mentioned in passing. If the author did not lean on it for the conclusion, attacking it does not weaken the conclusion.
A useful triage when you are stuck: ask which choice, if added to the stimulus, would most change the mind of a reasonable reader. The most mind-changing sentence is almost always the bridge attack, and the bridge attack is almost always the correct weaken answer. In my marking of practice sets, candidates who internalise this rule lift their CR accuracy by roughly 8 to 12 percentage points within three weeks, because they stop reading the choices as a buffet and start reading them as a competition to see which one targets the gap.
Strengthen questions: bolting the bridge back on
Strengthen items are the mirror image of weaken items. The stem asks which choice, if true, would most increase the force of the conclusion. The correct answer adds support to the bridge the author is relying on, or removes a competing explanation that would otherwise steal weight from the conclusion. As with weaken, the answer need not be a knockout proof of the conclusion. A modest but direct reinforcement of the bridge is enough, and it will usually beat a choice that is only tangentially related.
Strengthen items are where many candidates over-reach. They pick the choice that is most flattering to the conclusion in absolute terms, rather than the choice that is most directly relevant to the bridge. The right test is the same one you would use on a weaken item, but inverted: which choice, if added, would most change a reasonable reader's mind in favour of the argument. If two choices both feel relevant, the one that more precisely targets the bridge wins, even if the other choice sounds more dramatic.
Here is a worked shape. A retailer argues that the introduction of self-checkout kiosks caused average transaction time to fall by 18 percent. The bridge is that the kiosks caused the speedup, and that other factors were stable. A strengthen choice might confirm that staffing levels, store hours, and product mix were unchanged during the measurement period, isolating the kiosks as the cause. A weaker strengthen choice, often a distractor, would say that customers generally like self-checkout, which is true but does not isolate the cause. Train yourself to reject the likeable distractor in favour of the bridge-targeted choice.
One practical note: on the GMAT Focus, the same stimulus can appear with both a weaken stem and a strengthen stem in different forms, although never in the same test sitting. If you can solve a stimulus under one stem, mark the structure in your head so that when a future problem reuses a similar shape you can flip directions in under 30 seconds. This is one of the few short-cuts that survives scrutiny at the 84+ Verbal level.
The five trap answer patterns and how to spot each one
GMAT Critical Reasoning distractors are not random. They are written from a small library of patterns, and once you can name the pattern, you can eliminate the choice quickly. Here are the five I see most often, with a quick detection cue for each.
The bridge-adjacent decoy. This choice is true, plausible, and even supportive of the conclusion in some indirect way, but it does not address the specific bridge the question is testing. Detection cue: read the choice without the conclusion in mind, and ask whether it has anything to do with the gap you identified. If not, drop it.
The scope expander. The choice generalises a claim that the author kept narrow, or vice versa. The argument was about one city and the choice talks about the entire country. Detection cue: check the noun phrases. Does the choice's subject match the stimulus's subject, or has it been silently widened?
The opposite-but-related distractor. This is a favourite for weaken items. The choice is the negation of the correct answer, framed as if it were the correct answer. The author relies on X, and the choice says X is false. The trap is that on a weaken stem, negating the bridge is not the same as attacking it. The correct weaken choice usually gives a reason the bridge is weak, rather than asserting its opposite. Detection cue: if the choice is the literal negation of an assumption, be cautious; the GMAT prefers an explanatory weaken.
The evidence doubler. A choice that simply restates the evidence in different words, or adds a piece of evidence the author already used. It does not move the bridge. Detection cue: underline the bridge in your mind, then ask whether the choice changes the bridge. If the choice only thickens the evidence, it is decorative.
The plan-of-action distractor. Especially common on policy-style CR arguments, this choice suggests a solution the author never proposed. The argument is about whether a policy is justified, not about what to do next. Detection cue: if the choice says the city should or the company ought to and the stimulus only argued that a situation exists, the choice is out of scope.
Internalising these five takes about ten timed CR sets. After that, the trap answer usually shows itself inside the first 15 seconds of being read, which is what you want, because the GMAT Focus will not give you a quiet three minutes on any Verbal item.
Comparative table: how the three question types target the same skeleton
It is easier to read this section as a single table than to hold three separate mental models. The rows below show the three question families side by side, against the four parts of the argument skeleton, so you can see how each question type focuses on a different part of the same structure.
| Skeleton part | Assumption item asks | Weaken item asks | Strengthen item asks |
|---|---|---|---|
| Conclusion | What must be true for this conclusion to follow? | What makes this conclusion less likely? | What makes this conclusion more likely? |
| Evidence | Is the evidence the right kind, or only loosely related? | Is the evidence actually true or representative? | Is there extra evidence that confirms the original? |
| Bridge | What unstated step connects the two? | What alternative would steal weight from the bridge? | What fact would lock the bridge in place? |
| Scope or counterweight | Are there hidden limits the author ignored? | Is there a counter-example from outside the scope? | Is there a way to rule out the counterweight? |
When you can place the question stem in this table before you look at the choices, you have already done half the work. The other half is running the negation test on assumptions, the alternative-cause test on weakens, and the bridge-bolt test on strengthens.
Triage protocol for a CR item that refuses to yield
Even with a clean method, you will hit CR items on the GMAT Focus where the choices blur together and the clock is running. The protocol below is what I have students use when they have spent 60 seconds and have no strong lean. It is deliberately short, because most of the value of triage is in deciding to move on, not in heroic reasoning.
- Re-read the conclusion only. Five seconds. Confirm what is actually being claimed. In my experience, roughly 1 in 5 stuck items is stuck because the candidate has been arguing with the wrong conclusion for the previous 40 seconds.
- Re-read the bridge sentence you drafted. Five seconds. Confirm which gap the question is asking you to address.
- Rank the choices by bridge-relevance, not by language. Ten seconds. Discard anything that does not touch the bridge.
- Run the negation test on the top two. Ten seconds. Whichever answer becomes obviously broken when negated is more likely the correct answer on an assumption stem, and less likely the correct answer on a weaken stem, and so on.
- Pick the bridge-targeted choice and move on. Ten seconds. The expected point value of one more minute on a stuck CR item is lower than the expected point value of applying the freed time to your next Data Insights set or your next Quant item.
The 40-second budget above is realistic on the GMAT Focus, where the CR passages average around 90 words and the timing pressure is genuinely heavy. If a CR item takes you more than 1 minute 30 seconds in practice, the protocol is also your test-day decision rule, because sitting on a single Verbal item is one of the few moves that reliably destroys a 700-level total.
Common pitfalls and how to avoid them
The pitfalls below are the ones I see most often in CR error logs from candidates scoring in the 70 to 82 Verbal range. None of them are about content knowledge, and that is the point. CR is a method section, and the errors are method errors.
Reading the stem after the choices. The stem dictates the question type, and the question type dictates the bridge target. If you read the choices first, you will waste cognitive load comparing them on the wrong axis. Always read the stem, then re-read the conclusion, then look at the choices.
Treating strengthen and weaken as opposites. They are not symmetric. A weaken answer can be modest and still correct, but a strengthen answer usually needs to be targeted. Symmetry in your mind leads to picking distractors that are too strong on strengthens and too weak on weakens.
Letting world knowledge fill the bridge. The bridge is in the text, not in your head. If the stimulus says traffic fell and the conclusion says the new policy worked, the bridge is not people obey laws, it is no other factor changed traffic at the same time. Reject the answer that brings in outside knowledge.
Confusing the negation test for the falsifiability test. Negation does not mean the choice has to be obviously absurd. It means the bridge must collapse. A modest negation, like a small change in the sample, can be enough.
Skipping the conclusion in policy CR. Policy CR arguments often end with a recommendation, and the conclusion is the recommendation, not the descriptive sentence just before it. Candidates attack the description and miss the recommendation, then pick a choice that is true but irrelevant.
Building a GMAT Critical Reasoning practice routine that actually moves the score
Method without reps does not move scores. Below is the routine I assign to candidates aiming for V84 or above. It is built around 25 to 30 timed items per day, six days a week, with a structured review pass on the seventh.
Daily block, 45 minutes. Twenty-five CR items from the Official Guide or a reputable source, timed at 1 minute 45 seconds each. Mark the stem type before reading the choices, write the conclusion in your own words, draft the bridge, then run the negation test on assumption stems. Tag every missed item with the trap pattern that caught you, using the five names above. Do not re-do missed items on the same day; the short-term memory advantage is fake and will inflate your accuracy in a way that does not transfer to the real test.
Weekly review block, 90 minutes on day 7. Re-read each missed item, re-draw the bridge, and write one sentence explaining why each trap answer was tempting. The act of articulating the temptation is what moves the pattern into long-term memory. After the review, sort the missed items by trap pattern. If more than 40 percent of your misses come from one pattern, the next week's daily block should be built from items selected to expose that pattern. This is also where a private tutor or a structured Verbal programme can save you 20 to 30 hours of unfocused drilling.
Full-length rehearsal every 10 to 14 days. Take a full-length GMAT Focus practice test under timed conditions. Treat CR the way you will on test day. Do not pause to re-explain the bridge to yourself; just run the protocol. The point of the rehearsal is to make the protocol automatic, not to learn new material.
Score-report-driven retraining. After each full-length, look at the CR accuracy broken down by question type, not just total Verbal. If your assumption accuracy is 78 percent but your weaken accuracy is 62 percent, your next week of drilling should be weaken-weighted. The score report will tell you which bridge you are failing to attack.
One practical note on study materials. The Official Guide remains the single most accurate source for CR stimulus shape, because the writers are the test makers. Question banks from reputable third parties are useful for volume, but they occasionally drift in stem phrasing. If a third-party item reads as if the stem is asking a different question than the choices are answering, drop it and find another item. The GMAT itself is much more consistent in that mapping.
Why Critical Reasoning is the highest-leverage Verbal section on the GMAT Focus
Reading Comprehension and the new sentence-correction-adjacent items on the GMAT Focus each have their own value, but CR is where disciplined method pays off most reliably. Reading Comprehension rewards breadth of reading and slow, careful passage mapping, which is hard to acquire in 12 weeks of prep. Sentence-correction-adjacent items reward grammatical pattern recognition, which is real but narrower. CR rewards a single transferable method, the bridge test, that you can run on a fresh stimulus with no prior knowledge and still arrive at the correct answer.
This is also why, in a 12-week GMAT prep plan, CR is the Verbal strand where I push for the earliest score gains, often within three to four weeks of starting. The assumption-weaken-strengthen tripod is small, the trap patterns are few, and the transfer rate to Reading Comprehension is real: the bridge test is essentially the same skill that makes RC inference and RC tone items easier, because RC arguments also have bridges, just inside longer passages. Candidates who build the CR method first typically find their RC accuracy lift as a side effect.
For candidates balancing prep across Quant, Data Insights, and Verbal, that leverage matters. A 6 to 8 point lift on Verbal from focused CR work is realistic inside a 10-week window for a candidate starting in the V72 to V78 range. The same 6 to 8 point lift in Reading Comprehension often takes longer, because RC depends on more moving parts. If your score report says CR is your weakest Verbal family, that is not just a CR problem, it is the easiest Verbal problem to fix.
Conclusion and next steps
GMAT Critical Reasoning is a method section disguised as a reading section. The argument skeleton, the bridge test, and the five trap patterns are enough to solve the bulk of assumption, weaken, and strengthen items at a high accuracy rate, provided you rehearse them under timed conditions and tag your misses carefully. The biggest score lift usually comes from the triage protocol, not from new theory: the discipline of running the negation test on every assumption candidate, the alternative-cause test on every weaken candidate, and the bridge-bolt test on every strengthen candidate, inside 90 seconds, is what separates a V78 from a V84.
For candidates building a preparation plan around this Verbal strand, the next concrete step is a single timed set of 25 CR items drawn from the Official Guide, with a written bridge sentence for each, followed by a tag-and-review pass on the misses. TestPrep İstanbul's diagnostic assessment is a natural starting point for candidates who want their CR baseline mapped before they commit to a 10 to 12 week study plan.