A student who reads half a page of text and then asks "so what is actually being asked here" has already done most of the work. In scenario-based questions the difficulty does not sit in the topic knowledge; it sits in pulling the useful part out of the text. When someone who knows the material still cannot answer, this is usually why.
The format is everywhere now: in national school exams, in reading and problem sections of university entrance tests, in the open-ended parts of school papers. A situation is described, the data is scattered through it, and one thing is wanted.
What is a scenario-based question?
One that places the knowledge inside a situation instead of asking for it directly. "What is the formula for the area of a triangle" is a knowledge question. "A gardener sows a triangular plot of these dimensions, one kilogram of seed covers this much ground, how much seed is needed" is a scenario question built on the same formula. The second requires the first, but knowing the first is not enough.
School exams increasingly build this on purpose. Teachers pick a scenario first and write the open-ended questions around its learning outcomes, which means the text in front of you is not decoration, it is the frame of the question.
Why is the text so long?
Because the thing being measured is use, not recall. A short question cannot separate a student who memorised a formula from one who understands what it is for; both answer correctly. Put the same content in a situation and the difference shows immediately.
There is a side effect. Reading speed and attention now matter as much as subject knowledge. Read the question twice and you lose time; read it once and miss something and you solve the wrong problem. So the technique starts with changing the order in which you read.
Do you read the text or the question first?
The question first. Entering the text knowing what you are looking for is faster and safer than reading everything and then going back to find out what was wanted. The instruction usually sits in the last line under the text, often in bold or starting with "according to this".
The order works like this. Read the instruction, hold the thing being asked in your head as a single word (how many kilograms, which student, how many days), then read the text marking only the numbers and phrases that serve that word. You do not have to understand all of it. You have to understand enough of it to produce the answer.
Literature and reading passages are the exception. When a question asks what a poem or a paragraph means, the piece has to be read whole, because what is wanted is a judgement about the text rather than a single figure.
How do you strip out data you do not need?
By labelling every number as you meet it. Distractor data is placed deliberately: of five numbers in the question, three belong in the solution and two do not. A student who leaves the numbers bare and starts calculating uses all five and lands on a confident wrong answer.
A rule of thumb: data with no relation to the unit asked for is probably a distractor. If the answer is in kilograms, "four years ago" is usually part of the story. Not always, which is exactly why the sorting has to happen after you read the instruction and before you start solving.
In maths and science I would add one step: write the givens as a short list. Three lines turn half a page of prose into a solvable problem. Students who build this habit mostly stop saying "I did not understand the question", and what is left are genuine content gaps, which studying can close. The other reasons for that gap are in this piece on understanding a topic but not solving problems.
What about tables and graphs?
Read the headers. Most table questions catch students who skip the row and column labels. Which column is which year, which row is which product, is the unit thousands or millions. Work done without reading those can be arithmetically perfect and still answer the wrong question.
In a graph it is the axes and the scale. Whether the vertical axis starts at zero, whether the steps go up by two or by five, changes what the picture says. Read the scale first, then look at the shape, because an eye that does it the other way round sees a steeper slope than the data contains.
What if several questions share one text?
Read it once and take notes, rather than rereading per question. In clusters of two or three, most students go back to the text three times and spend half their time reading. A four-word summary in the margin makes each return quick.
Clusters carry another trap: the second question looks as if it depends on the first, and usually it does not. Getting the first one wrong rarely costs you the second, so there is no reason to panic through the rest of the group.
How do you practise this?
By rewriting questions before solving them. Take a scenario question and cut it down to two sentences, the givens and the thing asked for. Three a day for two weeks changes how you read.
The second method is writing your own. Take a plain knowledge question and build a situation around it; doing that once shows you where question writers hide data, and distractors get easier to spot afterwards.
For the reasoning side, our reasoning tests and puzzle games train the same muscle, moving from given information to a conclusion, and how to solve logic puzzles sets out that method step by step.