6 topics. Each one opens with its key idea, then the detail you need to use it.
1. Main idea and central idea questions
Key idea: The main idea is the claim the details serve; reject choices that cover one detail or claim more than the passage does.
The stem asks you which choice best sums up the text's main idea. The main idea, also called the central idea, is the one claim that every sentence in the passage serves. To find it, read the whole passage and ask what the details are there to prove, paying close attention to the first and last sentences. Writers often state the point up front, with a signal such as a claim that the benefits go beyond the obvious one, or they sum it up at the end with a conclusion.
Then test each choice with two questions: does it cover the whole passage or only one detail, and does it claim more than the passage does? Suppose a passage explains that a desert beetle's shell has water-attracting bumps that catch fog droplets, and that engineers are copying the pattern for fog-catching nets. A choice about the beetle alone is too narrow, and a choice saying engineers have solved water shortages is too broad. The right answer links the natural feature to the human use, at the same strength the text does.
A good main idea answer usually has two parts that match the passage's two parts. Often that is a phenomenon plus its effect, or a finding plus its meaning. If a choice delivers only one half, keep looking. Main purpose questions are close relatives, but they ask why the author wrote the text, such as to explain a process or challenge a theory, rather than what the text says. On the official PSAT, purpose questions belong to a different domain, Craft and Structure (Text Structure and Purpose), so they don't count toward this unit's share of the test.
2. Specific detail questions
Key idea: Locate the sentence, then pick the choice that restates its meaning and mechanism; any cause the passage doesn't state is wrong.
These stems point you to one part of the text: "According to the text...", "The text indicates that X happens because...", or "Based on the text, how does X work?" Specific detail questions are retrieval tasks, and the answer is a paraphrase of one or two sentences in the passage. Your job is to find those sentences and match the meaning, not the vocabulary.
The most common question asks for a mechanism or a cause. Take a passage about a titanium dioxide coating that breaks down pollutants when ultraviolet light hits it. If the stem asks how the coating removes them, the answer must describe a chemical reaction powered by light. A choice about filtering molecules by size sounds scientific, but it describes a physical process the passage never mentions. Ask yourself whether each choice uses the same mechanism as the text, not just the same nouns.
"Because" stems test a cause-and-effect chain that the passage spells out. A passage might say printed books got cheaper and literacy then rose, and the right answer will state that link. Wrong answers bring in plausible extra causes, such as schools that someone funded or a rule that forced people to read. None of these appear in the text, so none can be the reason the text gives. When two choices both seem right, return to the passage and look for the exact sentence each one paraphrases, because usually only one of them has a source.
3. Command of evidence: textual
Key idea: Split the claim into cause, effect and group; the right finding matches all three, and anything that shifts one part is wrong.
Command of evidence textual questions give you a claim and ask which finding, assuming it were true, would back the claim up most directly. Some ask which would weaken it instead. Because you assume each finding is true, the question is not whether a choice is true but whether it bears on this exact claim.
The method is to break the claim into parts before you look at the choices. Take a hypothesis that appearance-focused social media is particularly harmful to adolescent self-esteem. It has three parts: a cause (appearance-focused content), an effect (lower self-esteem) and a group (adolescents). The supporting evidence must keep all three parts intact. A study of adults changes the group, and a comparison of messaging and news apps never isolates appearance. A fact about the survey's reliability is about measurement, not about the hypothesis. The winning choice shows teens who left image-based platforms and then gained self-esteem. It changes the exact cause and tracks the exact effect in the exact group.
The strongest evidence often looks like an experiment, where one thing changes, the outcome moves, and everything else stays fixed. If the claim names a mechanism, such as sleep moving memories from the hippocampus to the neocortex, the best support shows that mechanism directly. A brain scan of reduced transfer in sleep-deprived students beats a general finding that tired students score lower. Some versions use literature and ask which quotation from a poem or novel best illustrates a claim about a character. The same rule applies: the quotation must show the exact trait the claim names.
4. Command of evidence: quantitative
Key idea: Match each number to its group, its unit and its direction, then confirm it addresses the claim the stem names.
Command of evidence quantitative questions put numbers in front of you. Sometimes the numbers are in the passage, and sometimes they are in a table or graph. The stem asks which choice describes the data correctly, or which choice uses the data to complete an example or support a claim. These questions reward slow, literal reading far more than they reward arithmetic.
Check every choice for three things before you commit to it. First, the right group: is the number about suburbs or downtown, high canopy or low canopy, the treated group or the control? Second, the right measure: is it degrees, percentage points or percent change? Third, the right direction: did the value rise or fall, and which group came out higher? Picture a passage saying cities run 2.9°C warmer than the countryside, and cities with lots of trees have heat islands about 40 percent weaker. A wrong answer will say cities are 40 percent hotter, and another will flip the trees so that bare cities are cooler. Both use genuine numbers from the text, placed in the wrong slots.
When the stem ties the data to a claim, read the claim first and then find the data that speaks to it. A table may show four years and three regions, but a claim about which region grew fastest needs only the start and end values for each region, compared across regions. You can ignore the other years and the extra detail. A choice can quote the table perfectly and still be wrong because it answers a different question. Also watch for claims about extremes and overall trends. "Rose every year" is false if one year dipped, even when the overall trend is up.
5. Inference questions
Key idea: Pick the conclusion the stated facts best support with no extra assumptions; hedged answers are often right, but nothing may go beyond the facts.
Inference questions end the passage with a blank and ask you to pick the most logical way to complete the text. You are finishing the author's argument, so the answer is the conclusion the stated facts best support, with no extra assumptions. Correct answers are often hedged with words like "probably" or "may", but they never go beyond what the facts support.
Work it like a chain by listing, in your head or on scratch paper, the facts the passage sets up. For example, a passage might say a lake fish only spawns in water below 12°C. It then says the lake has stayed above 14°C every spring since 2015. The logical completion is that the fish probably hasn't spawned there in recent springs. A choice saying the fish will go extinct goes too far, because the passage says nothing about other lakes or other breeding options. A choice about why the lake warmed answers a question nobody asked.
Watch the signal words just before the blank, because they tell you what kind of sentence belongs there. "Therefore" wants a conclusion, "however" wants a contrast with the previous sentence, and "this suggests that" wants an explanation of a pattern. The hardest inference questions set up two possible explanations and then give a fact that rules one out. The correct answer names whichever explanation is still standing. Test your choice by reading the whole passage with it in place; if you have to add an unstated assumption to make it work, it isn't the answer.
6. Informational text across the sciences
Key idea: Science passages explain a process or term; social science passages report a study; in studies, never upgrade a link into a proven cause.
Almost every passage in this unit is an informational text: it reports a finding and explains it. The field changes, but each field has a typical shape, and recognizing that shape speeds you up considerably.
Biology, ecology and environmental science passages usually describe a natural process step by step. Coral expels its algae when water warms, or plants release airborne chemicals that warn their neighbours. Questions usually ask what causes what, so track the order of events carefully. Physics and chemistry passages define a term, like gravitational lensing or photocatalysis, and then show a use for it. The definition sentence is often half of the main idea. The passage explains every technical word you need, so unfamiliar vocabulary is no reason to panic.
Economics, sociology, psychology and public health passages usually report a study. A research team surveys hundreds of teenagers or tracks rents across dozens of cities and then reports a pattern. Here the key distinction is between correlation and cause. Correlation means two variables tend to change together, so one tends to rise or fall as the other does, but it does not show that one causes the other. If neighbourhoods with gardens have less property crime, the study shows a link, not proof that the gardens caused the drop. Wrong answers often upgrade a correlation to a cause or add a word like "significantly" that the text never used. These passages also favor one contrast, in which a classic theory predicted one result and new data show something messier. When you see that set-up, the main idea is usually the messier result.