Sources and review

Smell mostly skips the brain's usual relay. The primary pathway reaches cortex without an obligatory thalamic relay, but the thalamus still has a role. Emotion and memory connections do not imply a direct route to the hippocampus.

In an exploratory fMRI study of five women, personally meaningful odours produced greater amygdala activation than comparison cues. This does not establish that perfume improves memory.

Among 93 older adults, odour-cued memories peaked before age 10, while word and picture cues peaked at ages 11–20. This was not a study of Hong Kong youth.

A study found confidence and video-rating differences after a deodorant containing fragrance and antimicrobial agents. Photographs showed no group difference; fragrance's isolated effect was not established.

Humans have about 400 intact olfactory-receptor genes, with substantial functional variation. A preference kit does not identify genotype.

OR7D4 variants influence the intensity and pleasantness of androstenone. They do not fully determine perception or explain all perfume preferences.

Skin is usually acidic. A study of 330 forearms found mean pH 4.93 after 24 hours without washing or products. The range 4.5–5.5 is not a universal healthy cutoff.

Carbon dioxide can bring unbuffered water near pH 5–6 at equilibrium. This does not establish a universal rate or a calibrated skin-test buffer.

Sniffin' Sticks uses blindfolded three-alternative forced choice and staircase reversals. Our simplified ladder is inspired by it, not validated as equivalent.

Some people have very low sensitivity to particular musk molecules. This does not mean they cannot smell every musk, or establish a diagnosis from the kit.

An exploratory fragrance-identification experiment found no coffee advantage over air or lemon. It measured novelty identification, not threshold recovery, and did not prove equivalence or harm.

The 57%/62%/86% figures come from a secondary blog account with conflicting significance wording. The full-paper statistics were not independently verified; omit a definitive lemon result.

A study of 14 volunteers across 120 tests reported mean recovery of 170 seconds, range 81–750, after saturating phenethyl alcohol. The authors recommended 15 minutes before retesting that odorant.

A pause ends exposure and allows recovery, but recovery depends on the odour, exposure and person. Our roughly 30-second pause is a workflow choice, not a proven complete reset.

DLG panel-training guidance lists coffee, lemon and elbow sniffing, excludes lemon in citrus/fragrance testing, and recommends 7–9 samples in its training exercise. Guidance is not a controlled efficacy trial.

Our bounded review found one exploratory coffee comparison and guidance on elbow sniffing. Comparative efficacy for skin, wool, water, charcoal or ceramic remains unverified; missing studies do not prove absence.

Hong Kong outdoor humidity normals are 78% annually and 81–83% in March–August. Sorption and mould studies justify packaging tests, but do not establish humidity, contamination or shelf life inside this kit.

Coffee storage research concerns freshness after package opening. Cited CO₂ and thiol figures were not independently confirmed and do not measure aroma migration into a kit or define its expiry.

The cited older IFRA standard concerns total bergapten under specific UV-exposure and measurement conditions. It does not establish current compliance for every citrus oil or product.

An empty clean vessel can mark a pause, but can acquire perfume residue and glass can break. Consumer-goods safety obligations still apply, with bilingual safety warnings where needed.

Classification of a coffee prop intended for smelling as food remains unverified. Applicable labelling and exemptions would need checking; a non-consumption label does not settle classification.

Describe a pause without promising restored smell. Whether a representation is an offence depends on facts; copy changes do not guarantee legal compliance.

The claimed 2024 Hong Kong fragrance-market total and growth could not be independently retrieved. Omit exact figures until dated extracts, definitions and exchange-rate assumptions are available.

An issuer reports 91.56% perfume use and 60.46% selection by mood, occasion or season among 1,063 Hong Kong–Macau respondents. Recruitment, weighting and youth breakdown were not disclosed; these are not population rates.

The notes' roughly 0.89 million estimate for ages 18–30 interpolates unverified five-year age bands. It is not an official count for that age range or a buyer estimate.

We have not identified an exact match in the reviewed sources. Custom blends, workshops and quiz-selected kits exist; marketplaces, imports and emerging sellers remain coverage gaps.

The reviewed Noteworthy snapshot lists a US$25 kit with US$25 bottle credit and four 8 ml vials, and n,551 100 ml at US$139. Earlier vial sizes and bottle prices were incorrect for that snapshot; refresh before comparing.

The reviewed direct-shipping policy lists the US and Canada. That does not establish absence of Hong Kong resellers, forwarding or import access.

Reviewed workshop examples differ in format: GLAMMI listed HK$460 for 30 ml/1.5 hours; HKU SPACE advertised HK$1,200 in September 2025. Selected examples do not bound the market.

Compare dated SKUs with matching volume and concentration. A reviewed Dior listing showed HK$1,150/100 ml and HK$850/60 ml; cached examples are not fresh quotes or category bounds.

The cited Le Labo prices are September 2023 examples; city exclusives exceed the proposed range and the Aesop figure was unverified. Refresh named prices before comparison.

Discovery HK$15–30, Lab HK$60–130 and 50 ml EDP HK$50–155 are preliminary goods-cost estimates, not Hong Kong supplier quotes. Labour, QC, delivery and one-off blending need quotes.

The cited 69.4% margin is a median of three public beauty firms, not a representative Hong Kong indie-fragrance sample or an Element forecast. Gross margin excludes contribution costs.

Under HK$4 is a selected token-cost scenario, not a ceiling. Images, retries, hosting and regional model prices are unmeasured; some listed model scenarios exceed it.

Carrier acceptance for our formula, volume, route and account remains unresolved. Retailer perfume deliveries do not authorise ours; written acceptance or a pickup plan is needed.

With full kit credit, kit price K plus bottle payment B−K equals B total revenue. Both goods costs, deliveries, fees, returns and acquisition must be subtracted; an existing offer does not prove Element's profit.

A ten-person laboratory study links some skin properties with evaporation; pH fell below its chosen importance cutoff. It does not validate consumer strips, photos or tuning. Benefit over a quiz-based approach remains to be tested. pH is recorded but never changes the formula.

Keller & Vosshall 2016: 55 people rated 480 molecules at two concentrations. The dataset is CC BY 4.0. In leave-one-person-out testing, Element's family-liking mean absolute error was 0.178 against 0.300 for a predict-the-average baseline, with k=3 nose types (Discerning, Open, Balanced). Transfer to blended perfumes is untested.