IB Mathematics Analysis and Approaches HL (AA HL) asks for two almost opposite things. Some questions demand a sealed proof, where every line must hold up on its own. Others invite you to probe a problem with a calculator, spot a pattern, and build a general argument from it. Revising topic by topic builds content knowledge, but it doesn’t train you to recognize which of the two a question wants. The result is a student who knows the chapter cold and still writes the wrong kind of answer.
Before any working, ask what would count as a complete answer. Algebra or a calculator on the page doesn’t settle it. Questions that say “prove,” “show that,” “for all,” “deduce,” or “disprove” demand a logically guaranteed result, so checking a few cases never finishes the job. Questions that say “investigate,” “find a pattern,” or “form a conjecture” signal exploration. Numbers are only the early phase there, and the answer stays incomplete until a general statement has been justified.
Paper 3 begins in exploration and then switches to proof language in its later phases. Expect that switch once the question asks you to state a result in general terms and explain why it must hold. That is the moment a calculator stops being enough.
Paper 3 as a Distinct Exam Demand
Paper 3 is a separate paper with its own format, its own demands, and a weighting to match. According to LiveTutor’s practitioner guide, only AA HL students sit it. It lasts one hour, is marked out of 55, and carries 20% of the HL grade. It consists of two compulsory extended problem-solving questions, and a graphic display calculator (GDC) is required. A paper worth a fifth of the grade and built around extended investigation isn’t a bonus to squeeze in after topic revision. It needs its own preparation logic.
The current Paper 3 format has been in place only since the 2021 syllabus change. That leaves a relatively small pool of papers in the current style. LiveTutor’s first recommendation is to use long investigation-style questions from pre-2021 HL papers, along with former exploration-style tasks, for extra practice. Its second is to save the specimen paper and the post-2021 Paper 3 papers for timed rehearsal. Practice material is the easy part. The harder part is knowing what a live question will ask of you once you open it.
How a Paper 3 Question Actually Unfolds
Pietro Meloni’s practitioner analysis describes a typical Paper 3 question as moving through four phases: compute specific cases, spot a pattern, state a precise conjecture, then prove and apply it. His worked example makes those phases concrete, using the sum Sₙ = ∑ₖ₌₁ⁿ k·k!.
In the first phase, students compute S₁ to S₄. In the second, they notice that adding 1 to each result gives a factorial. The third phase turns that observation into a conjecture, Sₙ = (n + 1)! − 1, which they check against the early values. In the fourth, they prove the conjecture by induction, then use the proven formula with a GDC to find the smallest n for which Sₙ exceeds 10⁶.
Meloni adds that exam questions often reward recognizing a different way of proving the same result in later parts. That alternative route is a telescoping sum, in which most of the terms in a long sum cancel against their neighbors and leave only a few.
The four phases make two different demands. The first two are exploratory, driven by the GDC and numerical experimentation. The last two require formal rigor: a precisely stated conjecture and a logically complete proof. Students who treat the whole question as open-ended exploration under-deliver on that proof expectation.
The example shows where the line falls. Computing S₁ to S₄ is case-checking, however many cases you add. The work becomes proof when the induction begins, and the calculator comes back only at the end, to apply a formula that is already secure.
The Proof Demands of Papers 1 and 2
The habits that make Paper 3 go well—numerical experimentation and conjecture-building—don’t directly match what handwritten proof demands. Paper 1 is the strictly no-calculator paper, but the sustained proof expectations belong to both Paper 1 and Paper 2. Handwritten proof rewards three habits: holding the structure of the argument in your head, committing to an approach before you can verify it, and writing each justification so it stands on its own.
Without specific practice, students tend to write proofs that are roughly right in direction but skip justification steps they would normally rely on a calculator to check.
One valid counterexample is enough to disprove a universal claim, but only if it is presented cleanly, with the factorization shown explicitly. A worked HL AA proof question from IB Math Revision shows what that looks like. The claim is that 2ⁿ − 1 is prime for every prime n. The worked solution checks prime values systematically until n = 11, where 2¹¹ − 1 turns out to be composite, and the explicit factorization of that number completes the disproof.
The same resource treats induction as structure-heavy writing: a base case, an inductive hypothesis, an inductive step, and an explicit conclusion. That matters because comparing your writing with a worked solution is where the gaps show.
Using Past-Paper Solutions to Sharpen Mode Recognition
Past-paper solutions do their best work after you’ve attempted the question under real exam conditions, not in place of the attempt. That means no calculator for Paper 1, a GDC for Paper 3, and extended time for investigation-style questions. Working in that order keeps your independent reasoning intact and gives you a benchmark for what each kind of question expects. Maths AA HL past-paper solutions, whether written markschemes or step-by-step video walkthroughs such as those from Revision Village, show the reasoning structure expected at each stage, so you can set it beside your own.
For Paper 3, the aim of the check is to see whether you used the right type of reasoning at each stage. First, find the point where the solution stops listing specific cases and states a general conjecture. Your own work should contain an explicit general statement, not just a table. Second, find where the solution switches from numerical evidence to a proof argument that would still work without a calculator, such as an induction or telescoping proof. Third, note where the GDC appears: after the general result is established, for applications or threshold-finding. That tells you whether you used the technology in the right phase.
For Paper 1 and Paper 2 proofs, compare structure rather than algebra. Find the line where the solution explicitly justifies each major claim, such as the explicit factorization that completes a counterexample disproof, or the base case, inductive hypothesis, inductive step, and conclusion in an induction. Then check whether your writing contains those moves or merely implies them. Make either comparison straight after the attempt, while your choices are fresh. A missing explicit justification step or a missing general statement is a signal that the mode wasn’t fully met, even if the final answer matched.
Integrating Proof and Investigation in AA HL Prep
The two modes aren’t rivals. They strengthen each other. Practicing the Paper 3 phase pattern teaches you how an extended question unfolds. Exploring carefully at the start gives your conjectures something sound to rest on. Writing proofs under Paper 1 conditions teaches you to put the reasons on the page rather than keep them in your head.
AA HL will keep handing you both the question that wants a sealed proof and the one that invites you to probe, and sometimes both in the same problem. So organize revision by reasoning mode as well as by topic, and use past-paper review to close the gap between what you wrote and what the question was asking for. Skip that, and you can know the chapter thoroughly and still reach for the wrong mode.
