Topic Brief: Hidden Higher Degree Equations From Differentiation

Provenance: compiled 2026-09-29, from research/veridian/WMA11-verified-facts.md §4.4 and its "Lesson-audit log" entry for hidden-higher-degree-equations-from-differentiation.ts (2026-09-14) — both already re-extracted from real QP/MS/ER PDFs. No fresh PDF extraction was done here. signal-frequency-and-boundaries.md §1 rolls this topic into a broad "Differentiation — 4 of 5" row with no sub-topic breakdown, so this is a new drill-down, not an expansion of an existing line. Examiner-report material is paraphrased rather than quoted at length; the one short quotation is attributed. Not affiliated with or endorsed by Pearson Edexcel.

#SIGNAL Topic Master Brief — Hidden Higher-Degree Equations From Differentiation (spec 4.2, applications via 4.1/4.3)

#Frequency: seen in 2 of 5 series checked

The facts bank fully mined five WMA11 examiner reports question-by-question: Jan 2019, Oct 2020, Oct 2021, Jan 2023, Jan 2025. This exact topic — differentiate, set the result equal to a target, and end up with a quartic that is actually a disguised quadratic in x² — is confirmed in 2 of those 5: Oct 2021 (Q2) and Jan 2025 (Q5(c)). It did not appear in Oct 2020's fully-read report; Jan 2023's own differentiation question (Q1) tests a different skill (differentiating negative/fractional indices to an exact gradient, with no equation to solve); Jan 2019's own differentiation question (Q6) was checked directly and also confirmed to test a different skill — see "Checked and ruled out" below.

Two further series (Oct 2022, Jan 2024) were confirmed genuine but never mined question-by-question, and two June series (June 2022, June 2023) were read for one targeted question each, neither a differentiation question — so "2 of 5" is honest only for the five series read in full. This sits below the broader "Differentiation: 4 of 5" row elsewhere in this folder, which counts any differentiation question at all, not this specific quartic-in-disguise pattern.

#The two confirmed appearances

Oct 2021 Q2 — curve y = 3x⁵ + 4x³ − x + 5; gradient equals 2 at both P and Q; find the x-coordinates. 5 marks, a single unsplit part: M1 (differentiate, one correct power) → A1 (dy/dx = 15x⁴+12x²−1) → dM1 (set equal to 2, collect to a 3TQ in x², dependent on the first M) → ddM1 (solve the 3TQ, dependent on both prior method marks) → A1 (x = ±1/√5, both values required).

Jan 2025 Q5(c) — curve y = 4x³ + 2/x + 9 (x > 0); part (b) shows 12x⁴+5x²−2=0; part (c) finds k where a line y=k−5x is tangent at P. Part (c)'s 4 marks: M1 (solve the given quadratic-in-x² for a positive x² and proceed to a positive x) → M1 (substitute that x into the curve's own equation for y) → dM1 (use the x,y pair and the tangent line to find k) → A1 (k = 16 only).

#Two different route-ins to the same disguised quadratic

The two confirmed anchors reach an "x⁴ and x² only" equation by different paths. Oct 2021 Q2's f(x) has no negative-power term at all — a plain degree-5 polynomial whose derivative (15x⁴+12x²−1) is already a quartic, no fraction-clearing needed. Jan 2025 Q5's curve carries a negative-power term (2/x), so its quartic only appears once the fraction is cleared by multiplying through by x². Both routes end at the identical skill — no x³ or x¹ term survives, substitute u=x², solve the resulting 3TQ — but one confirmed series reaches it by differentiating a higher-degree polynomial directly, the other by clearing a negative index. Expect either.

#Mark-tariff pattern

The two confirmed sightings differ sharply in how self-contained the skill is. Oct 2021 Q2 is the whole question — 5 marks, start to finish. Jan 2025's version sits inside a longer tangent-line problem: forming the quartic is a "show that" step in part (b), and the disguised-quadratic solve is only the first of four marks in part (c), alongside a substitution step, a further dependent method mark, and a final accuracy mark belonging to the surrounding tangent/k-finding context, not the quartic itself. Don't assume this topic always carries its own clean 4–5 mark block — it can just as easily be one M1 inside a longer curve-and-line question.

#The recurring angle: domain restriction is inconsistent, and that's the real trap

The two confirmed appearances disagree on whether both signs of x survive. Oct 2021 Q2 has no domain restriction: its one valid root of u (1/5, after rejecting −1) converts to a genuine pair, x=±1/√5. Jan 2025 Q5 opens with "x > 0" stated up front, so only the positive branch is a valid final answer — and the same examiner report records candidates who kept both signs anyway losing the final mark for it. The pattern across both: check the question's own opening line for a stated domain before defaulting to ±√u, because whether that default even applies is not consistent from one sitting to the next.

#Confirmed trap set (both series, examiner-report-sourced)

• Quartic not recognised as solvable at all. Oct 2021 Q2's examiner report notes some candidates saw no route into solving the resulting quartic.
• Substituted variable confused with the original. The same report records a small number who replaced x² with x rather than the reverse, rarely reaching a correct answer.
• Factorisation doesn't match the unsimplified equation. Also Oct 2021 Q2: many lost a mark for a factorised form that didn't actually follow from dividing the unsimplified quartic through by 3 — a presentation/rigour mark, not an arithmetic one.
• Forgot to square-root back to x. In Jan 2025 Q5(c), candidates who correctly reached x²=¼ sometimes used ¼ itself as x, skipping the square root and scoring nothing on that step.
• Kept both signs despite a stated domain. The opposite error on the same question: substituting a negative x too, losing the final mark "despite 'x > 0' in the wording of the question" (Jan 2025 Q5(c) examiner report).

#Checked and ruled out: Jan 2019 Q6

An earlier research pass had listed Jan 2019 Q6 as a possible third anchor on a bare parenthetical. A direct re-fetch of the real QP/MS/ER (not in the local archive; fetched live) confirmed it tests something else: f(x) = 2x^(5/2) − 40x + 8, x > 0, where f'(x)=0 and f''(x)=5 both solve by raising a fractional-index term to a reciprocal power (e.g. x^(3/2)=8 → x=4) — never producing a polynomial in only even powers of x. The real examiner report's own diagnosis confirms the different skill: candidates struggled with reversing the index operation itself, not with recognising a disguised quadratic. Named here so a future pass doesn't re-inherit the miscount: this topic's real base rate is 2 of 5, not 3.

#If this topic appears: what to expect

Expect either a short, self-contained ~5-mark question (Oct 2021's shape) or a single dependent method mark inside a longer tangent/normal-in-context question (Jan 2025's shape) — both confirmed shapes are real. Either way: differentiate first (watch the sign on any negative-power term), set the result equal to the target, clear any fraction, and check whether only even powers of x survive. If so, substitute u=x², solve the ordinary 3TQ, reject any negative u outright, and square-root every surviving positive u to a pair of x-values — unless the question's own opening line already restricts the domain, in which case only the branch that satisfies it is valid. Show the division/simplification line explicitly before presenting a factorisation; a real mark scheme has already refused to credit one that doesn't visibly follow from the equation as written.