Cider science
Training to recognise faults
How do people learn to taste faults in cider?
In short
By tasting the fault on purpose, at a known strength, beside the same drink without it. A trainer adds a measured amount of the compound to a neutral base, tasters learn to pick it out, and the amount is reduced until they can only just tell. Training usually starts in water and moves into cider, because the drink itself hides and changes the fault.
Two thresholds matter, and they are different. The detection threshold is where someone can tell that something is there. The recognition threshold is higher: where they can say what it is.
Training sharpens both, but it has limits. Some people cannot smell particular compounds at all, and in a published wine study some taints were missed by a whole panel at the concentration usually quoted as the threshold.
Detection and recognition
ASTM E679, the standard method for measuring odour and taste thresholds, distinguishes the concentration at which a person becomes aware that something has been added from the concentration at which they recognise what it is. It measures them with forced choices, sample against blank, in an ascending series of concentrations, so that a lucky guess at a low level is not mistaken for detection.
The standard also warns that training can profoundly change the threshold a person shows, and that thresholds measured by one method are not necessarily equivalent to those measured by another. A published threshold is a property of the method and the panel as much as of the compound. It is a guide to where a fault starts to matter, not a line every drinker crosses at the same place.
How a training sequence is built
- Start with reference standards in water, so the compound is learned without the drink around it. One Virginia Tech enology guide describes training that begins with standards in water and continues with standards in the wine.
- Then move into the product. The drink changes thresholds: in the wine study described below, the matrix influenced what panellists could detect.
- Present each reference beside an untreated control, coded, and ask for a forced choice. A taster who is told which glass is spiked learns the label, not the smell.
- Work from clearly above threshold downwards. Recognition is learned at a strength that can be recognised.
- Use a single descriptor at a single intensity per reference. The cider lexicon study trained its panel this way, over 13 hours.
- Retest. A taster who recognised a fault last season has not necessarily kept the skill.
What a wine panel found at threshold
A 2018 study spiked wine with three taints, each at its published odour detection threshold: a smoke compound, cork taint (TCA) and the green-pepper pyrazine IBMP. Thirty-four panellists each did the triangle tests in triplicate. They distinguished the smoke-spiked wines from the controls, but not the cork-taint or pyrazine samples. Panellists with previous experience or training did better.
The point for a cider maker is not the wine numbers. A published threshold can be a concentration that a real panel, in a real drink, does not reliably detect. Training to recognise a fault means training at strengths people can actually perceive, then finding where each taster's own limit is.
Blind spots that training does not fix
Specific anosmia is the inability to smell one particular odorant despite an otherwise normal sense of smell. In a 2020 study of rotundone, the peppery compound in some red wines, roughly 40% of the people tested could not smell it at all, and the thresholds of the rest varied markedly.
For a cidery this means one person's nose is not a quality system. A cider maker who cannot smell a particular fault will pass every batch that has it, with complete confidence. Screening each taster against the faults that matter, and having more than one person check each batch, is the practical response.
Also answered on this page
Questions this page covers, so you can tell at a glance whether it is the one you want.
- How do you train someone to taste cider faults?
- What is the difference between detecting and recognising a taint?
- Why can some people not smell a fault others find obvious?
- Is a published taint threshold the level everyone notices?
Related
What people ask next
Questions readers ask about the things this page mentions. Each one goes to the section that answers it rather than to a page written to receive the question.
- Is that a fault or is it meant to taste like that?
- Something is wrong with my cider — what is it — Start from what you can actually observe — the smell, the taste, the appearance and the behaviour in the glass — rather than from a guess about the cause. The fault finder narrows a description down to the faults that match it.
- What is mousiness in cider? — Mousiness is a fault caused by tetrahydropyridines produced by *Brettanomyces* and some lactic acid bacteria. It tastes of stale grain or a mouse cage and appears in the aftertaste rather than in the aroma.
- What is cork taint? — Cork taint is a musty, damp-cardboard smell caused mainly by trichloroanisole, which can come from the cork or from the cellar. It strips aroma from the drink even below the level at which it can be identified.
- What is a sensory threshold? — A threshold is the concentration at which a compound becomes detectable. Thresholds vary by orders of magnitude between compounds and between people, which is why one taster finds a fault obvious and another cannot find it at all.
- What is smoke taint?
Sources
What this page rests on. Where a source is marked as registered rather than read, CiderHQ is recording that the body is authoritative on the subject without claiming to have worked through the document itself. See our evidence policy for what each state means.
ASTM International (public scope page) · standards body · passage verified 2026-10-29
How this source was checked
Only the public scope text was read; the standard itself is sold and was not bought.
McKay and colleagues, Foods 7(11):176 (2018), doi:10.3390/foods7110176 · peer-reviewed literature · passage verified 2026-10-29
How this source was checked
Wine, not cider. Read for how taint recognition is tested by spiking and how training changed detection.
Individual differences in thresholds and consumer preference for rotundone added to red wine
Gaby and colleagues, Nutrients 12(9):2522 (2020), doi:10.3390/nu12092522 · peer-reviewed literature · passage verified 2026-10-29
How this source was checked
Wine, not cider. Cited for the size of individual threshold variation and for specific anosmia.
Sensory Analysis, Sections 3 and 4 (winemaking issues series)
Bruce Zoecklein, Enology-Grape Chemistry Group, Virginia Tech · university · passage verified 2026-10-29
How this source was checked
Written for winery panels. Section 3 (panel training) and Section 4 (difference tests) were read.
Calvert, Neill, Stewart and Lahne, Journal of Food Science (2023), doi:10.1111/1750-3841.16507 (open access, Virginia Tech repository) · peer-reviewed literature · passage verified 2026-10-29
How this source was checked
A cider-specific trained descriptive panel. Read for its training and serving protocol.
ISO 4120:2021 Sensory analysis — Methodology — Triangle test
International Organization for Standardization (publisher’s preview via iTeh Standards) · standards body · passage verified 2026-10-29
How this source was checked
Preview of the standard, which includes the scope, the procedure and Annex A Table A.1 of critical values. The discrimination-test calculator is tested against that table row by row. The full standard was not bought.