Most "difficult" exam questions aren't actually difficult
Andy, PrepLab co-founder · 16 Nov 2026 · Updated 15 Nov 2026

This might sound strange, but I don't think most difficult exam questions are actually testing difficult concepts.
They're usually testing familiar concepts in unfamiliar ways.
The panic moment
When students see a hard question, the first reaction is often panic.
"I've never seen this before." "I don't know how to start." "This wasn't in the notes."
I used to think the same way. But after doing enough questions, I noticed something.
The hardest questions were almost never testing brand-new ideas. The concepts were usually things I'd already learnt. The examiner had just changed the context, combined a few topics, or asked from a different angle.
The idea underneath was still the same.
Hard questions sit a small distance from the theory
A difficult question isn't difficult because it comes from nowhere. It's difficult because it's slightly further from the most obvious use of the theory. But it's still connected to it.
The examiner is limited by the syllabus. The question has to be built on something you were taught, and it can't break the core theory.
So the answer must be somewhere near the theory. Your job is to find the path.
With a strong understanding, you can travel that distance.
The one question to ask
Instead of asking, "Have I seen this question before?"
I started asking, "Which idea is this question really testing?"
That small change made a huge difference. Because concepts are limited. Question styles are unlimited.
If your learning is based on recognising question types, every new question feels scary. If it's based on understanding concepts, unfamiliar questions become much easier to find your way through.
Panic comes from unfamiliarity. Understanding comes from recognising what's underneath.
There's research behind this idea. In a classic 1981 study in Cognitive Science, Michelene Chi, Paul Feltovich and Robert Glaser asked physics experts and beginners to sort physics problems. Beginners tended to group problems by what was literally in them. Experts grouped them by the physics principle needed. In the authors' words, "experts initially abstract physics principles to approach and solve a problem representation" [1].
In other words, experts ask "which idea is this testing?" first.
What it looks like in your subjects
E-Math. A ladder leaning against a wall. A kite on a string. The diagonal path across a field. Three different stories, one idea underneath: Pythagoras' theorem. The question just wraps it in a new picture.
A-Math. Water is poured into a cone at a steady rate. How fast is the water level rising? It looks scary. Underneath, it's connected rates of change: the chain rule (dh/dt = dh/dV × dV/dt) plus the formula for the volume of a cone.
Physics. A wheelbarrow, a spanner, a see-saw with two people on it. All of them are the principle of moments: when something is balanced, the clockwise moments equal the anticlockwise moments. The setup is new. The law isn't.
Chemistry. "Metal X displaces copper from copper(II) sulfate solution, but not zinc from zinc sulfate solution." You've never heard of metal X. It doesn't matter. It's the reactivity series, and X sits between zinc and copper. Or rubidium: you may never have studied it, but Group 1 trends tell you it reacts with water even more vigorously than potassium.
It's the same in other subjects. In Economics, a market failure question can look very different depending on the context, but the logic underneath stays the same.
The surface changes. The theory remains.
Practice should build understanding, not replace it
Practice is still important. But there's a danger in doing many questions without understanding.
You can get good at recognising question types but stay weak at reasoning. The moment a question looks different, the method disappears.
That isn't real mastery. Real mastery is when the concept is so clear that even an unfamiliar question can be broken down.
One way to practise this is to mix topics you've already learnt, instead of doing twenty of the same kind in a row. Researchers call it interleaving. In a US study by Doug Rohrer and colleagues, maths practice was "arranged so that no two consecutive problems can be solved by the same strategy". That "forces students to choose an appropriate strategy for each problem on the basis of the problem itself" [2].
Across 787 Grade 7 students in 54 classes, the group with more mixed practice scored 61% on a surprise test, against 38% for the others [2]. That's one study in one school district, but it's exactly the skill hard questions test: working out which idea you need before you start.
Simplify, but keep it true
To me, the goal was always to explain the idea in my own words.
Not simplified and wrong. Simplified and correct.
That's the difficult part. Anyone can simplify by leaving out important details. True understanding is being able to simplify while keeping the truth.
So for every topic, I asked myself:
- Basically, what does this mean?
- What's the simple truth here?
- What must never be broken?
- What does the examiner actually want to see?
Once those are answered, the subject becomes much more manageable. If your explanation still sounds like copied notes, the concept probably hasn't sunk in yet.
After the click
I think this is why some students suddenly improve very quickly.
Before the click, everything feels disconnected. Every chapter feels separate. Every question feels like another thing to memorise.
After the click, the subject begins to feel smaller. The same ideas keep appearing. The same logic keeps repeating.
At that point, learning speeds up. It's less about memorising more and more, and more about connecting what you already know.
A memorised answer is useful for one question. A concept is useful for many. A deeply understood concept stays useful long after the exam.
That's when difficult questions become solvable. Not always easy. But solvable. Because your mind has something solid to come back to.
So don't study only to recognise questions. Study to understand the idea behind them. The goal isn't to memorise every possible question. That's impossible. The goal is to understand the idea so well that even a new question still feels familiar.
If you use PrepLab
Every wrong answer in PrepLab is saved to your Mistakes, and you can review them by topic. Next time you go through them, don't just check the answer. Ask, "Which idea was this testing?" And once Standard questions feel easy, try the Advanced ones: that's where you practise travelling the distance from the theory.
Short daily sets with your friends, and every mistake saved to try again. Free to download on iPhone. Download on the App Store
Sources
- Chi, M. T. H., Feltovich, P. J., & Glaser, R. (1981). Categorization and representation of physics problems by experts and novices. Cognitive Science, 5(2), 121–152. https://doi.org/10.1207/s15516709cog0502_2 (abstract opened 2026-11-15)
- Rohrer, D., Dedrick, R. F., Hartwig, M. K., & Cheung, C.-N. (2020). A randomized controlled trial of interleaved mathematics practice. Journal of Educational Psychology, 112(1), 40–52. https://doi.org/10.1037/edu0000367 ; quotes and result from the IES award page "Efficacy Study of Interleaved Mathematics Practice", https://ies.ed.gov/use-work/awards/efficacy-study-interleaved-mathematics-practice (opened 2026-11-15)