'For all' (∀) is universal: it claims the property for every member of the domain — for all integers n, n² ≥ 0. One example never proves it, and weakening it to 'at least one' is the standard misread. 'There exists' (∃) promises just one witness: n = 2 settles 'some n has n³ = 8', and p = 2 settles 'for some prime p, p is even'. Ordinary words hide quantifiers — always, guarantees and ensures are universal; sometimes, can and may are existential.
Order is everything when quantifiers nest. 'For every integer a there is an integer b with b = a²' is true — the witness b = a² may depend on a. Reverse them ('there is a b such that for every a, b = a²') and one fixed b must equal every square, which is impossible since squares are unbounded. The same hinge decides the divisibility pair and the k + n parity worksheet: letting the witness depend on the earlier variable is allowed in ∀∃ and forbidden in ∃∀.
Traps: reading 'there exists' as 'exactly one' or 'for every', and assuming quantifiers commute. For 'must be true' questions, attack each option by constructing a case where every premise holds and the option fails — if you can build it, eliminate without mercy. And remember the asymmetry: confirming instances (3 × 4 = 12 is even) never establish a universal, while a single genuine counterexample destroys one — the bridge to negation and disproof. Before finalising, restate the claim aloud with 'every' or 'some' in place of the original wording — if the restatement feels stronger or weaker than the stem, you have caught the scope shift the distractor was counting on.
Study method: attempt five timed questions, log every miss with its topic code, redo misses from scratch within 48 hours, then re-attempt the topic a week later. Pair each hub with its drill link below and a fortnightly full mock.
Place this topic in the wider test with the syllabus checklist and Paper 1 vs Paper 2, learn the timing system in how to prepare, and check what scores mean in scores explained.