Instrumental colour values are computed through a declared standard observer. A real viewer can make a different match without making the instrument broken or the viewer defective.
This guide follows spectra, 2° and 10° colour-matching functions, observer metamerism and a small fictional viewing panel. It keeps model output, instrument uncertainty and human variation in separate records.
All spectra and responses are invented teaching cases. They do not characterise a person, artwork, pigment or commercial display.
The practical habit is to preserve a small evidence receipt beside every answer. It names the object, reference condition, units, stage or algorithm, observed inputs, transformation and the nearest tempting claim that still does not follow.
Each changed-context check alters one consequential condition while keeping the remaining basis fixed. This makes transfer visible: the reader can predict which field should change, which should remain stable and what new evidence would be required.
The article uses five routes so readers can enter by task, but the mechanism is cumulative. Definitions establish the objects, the worked case carries the arithmetic, evidence checks test rivals, transfer cases change the context and the final route repairs common shortcuts.
This specialist guide belongs to the Visual Arts root. It also connects to The Gold Standard of Education | Learning That Transfers into Life and the How X Works hub. The worked cases are original teaching examples; their numbers are stipulated so that the reasoning can be checked.
eduKateSG · Visual Arts
Find your next learning step
Read the short answer first: use a standard observer for reproducible colorimetry, then treat individual visual matches as separate evidence collected under controlled viewing conditions. Agreement under one does not guarantee agreement under the other.
ROUTE 1 · CHAPTERS 1–4
Define the quantities
Name the quantities, reference conditions and reporting rules.
ROUTE 3 · CHAPTERS 9–12
Audit the evidence
Check instruments, algorithms, assumptions and rival explanations.
ROUTE 5 · CHAPTERS 17–20
Report and continue
Repair shortcuts, answer questions and follow verified routes.
Chapter contents
Define the quantities · Chapters 1–4
1. Name the mathematical observer and the person
2. Choose the 2-degree or 10-degree reference
Work the mechanism · Chapters 5–8
5. Work the central display-pair case
6. Calculate tristimulus values in order
Audit the evidence · Chapters 9–12
9. Keep viewing field geometry visible
10. Control adaptation and background
Transfer the reasoning · Chapters 13–16
13. Transfer to narrow-primary displays
14. Transfer to pigments and proofs
Report and continue · Chapters 17–20
17. Repair instrument says everyone matches
18. Report panel variation honestly
CHAPTER 1 OF 20 · Define the quantities
1. Name the mathematical observer and the person
A CIE standard colorimetric observer is a specified set of colour-matching functions used to compute tristimulus values. An individual observer is a person whose visual match can differ from that reference even with normal colour vision.
Write the evidence in the order a reader can audit: the Visual Arts quantity, its reference object, the observation or browser rule, the transformation, and the decision being made. Keeping those fields visible prevents a familiar label from hiding a changed basis.
For two stipulated spectra, the standard-observer calculation gives equal XYZ values. Two participants then make separate visual judgements; their responses are observations, not recalculations of the same model.
First-principles test: ask what physical, observational or algorithmic operation produces each value in this chapter. Then ask whether the second value could change while the first stays fixed. If it can, the two labels are not substitutes. For Visual Arts, this counterfactual is often clearer than a definition learned in isolation.
Keep the spectra fixed and replace the standard function set with a measured individual function set. The computed match can change.
Evidence check: another reader should be able to reproduce the narrow result from the stored definitions, inputs and intermediate steps, then identify the nearest stronger claim that the evidence does not support.
This article does not diagnose colour-vision conditions or predict one person’s perception.
CHAPTER 2 OF 20 · Define the quantities
2. Choose the 2-degree or 10-degree reference
The CIE 1931 2° and CIE 1964 10° observers represent different field-size conditions. Selecting one is part of the measurement method, not a cosmetic metadata choice.
Trace the mechanism before interpreting the result. For Visual Arts, mark which value belongs to the source, which belongs to the measurement or layout algorithm, and which is a reported summary. A calculation is useful only when those roles remain attached.
A small central swatch is computed with the 2° observer; a large comparison field is recomputed with the 10° observer. Store both results with the field geometry.
Practical transfer: build two records that share the headline number but differ in the condition discussed here. If the correct interpretation changes, the omitted condition is decision-relevant and belongs in the report. This test turns a vocabulary distinction into a reusable Visual Arts checking habit.
Enlarge the stimulus while leaving spectra unchanged. The appropriate reference may change even before any participant data are collected.
Practical check: keep one counterexample beside the worked case. If the wording fails when that one condition changes, narrow the sentence until it describes only the evidence actually in hand.
Exact applicability also depends on adaptation and photopic viewing conditions.
Tristimulus values are integrations of spectral power or reflectance, illuminant and colour-matching functions. Equal coordinates under one observer do not prove equal spectra.
Use a compact decision table with columns for condition, input, operation, result and limit. This makes the comparison reproducible and exposes where a missing field would force an assumption instead of an observation.
Spectrum A uses a broad green band; spectrum B uses two narrower bands. They are stipulated to match under the reference observer while remaining spectrally different.
Misconception check: a plausible output does not prove the right reference, denominator, ancestor, spectrum or observation rule was used. Reconstruct one intermediate step and one changed-context prediction. If either cannot be stated, keep the result provisional rather than strengthening its wording.
Change the illuminant or observer functions and predict that the match may break.
Transfer check: repeat the reasoning with one consequential condition changed while holding the rest fixed. State what changes, what stays comparable and why the original label alone was insufficient.
A coordinate example without measured spectra remains an illustration.
CHAPTER 4 OF 20 · Define the quantities
4. Treat the standard observer as a convention
The standard observer supports reproducible colorimetry across instruments and organisations. It is representative under specified conditions, not a promise that every person will report the same match.
Preserve the raw inputs beside the derived result. If a later reader changes one condition, the original Visual Arts reasoning should still be recoverable without reverse-engineering a rounded headline or an unlabeled screenshot.
A spectrophotometer produces one standard-observer report for a proofing pair. A viewing panel records a spread of match judgements. Both datasets can be valid.
Evidence design: retain both a positive case and a near-miss. The positive case shows when the rule applies; the near-miss changes one consequential condition and shows when the conclusion fails. Together they teach the boundary of the Visual Arts mechanism better than another unlabelled example.
Repeat the panel with a new observer group while keeping the instrumental calculation fixed.
Audit check: retain the source version, units, settings, exclusions and rounding rule. A future correction should be able to alter one field without rebuilding the entire explanation from memory.
Panel design, screening and consent require an appropriate research protocol.
The central fictional case uses two display spectra P and Q that compute to the same reference XYZ values within rounding. Observer A calls them a match; Observer B reports a visible difference; Observer C is unsure.
Ask what would have to stay fixed for the comparison to be fair. Record those controls explicitly, then name the first uncontrolled condition that could reverse the conclusion. This is the practical transfer step, not an optional footnote.
Record the standard result as model-based equality under the named functions and the panel result as three individual judgements. Do not average them into one invented “true colour”.
Communication test: replace any sentence that says two things are “the same” or “different” with a sentence naming the operation, reference condition and supported consequence. The extra words are useful when they let a future reader repeat the decision without access to the original author.
Change only Q’s narrow red primary and recompute the standard difference before collecting new judgements.
Decision check: end with an action or interpretation that follows from the stated conditions, followed by a boundary sentence explaining when a specialist rule, local procedure or additional measurement is needed.
The values are stipulated and do not characterise any commercial display.
Multiply the stimulus spectrum by each colour-matching function and integrate over wavelength to obtain X, Y and Z on the declared normalization. The same spectrum can produce different coordinates under another observer function set.
Write the evidence in the order a reader can audit: the Visual Arts quantity, its reference object, the observation or browser rule, the transformation, and the decision being made. Keeping those fields visible prevents a familiar label from hiding a changed basis.
Use a five-band toy table so every product and sum is visible. Equal final triples can arise from different band contributions.
First-principles test: ask what physical, observational or algorithmic operation produces each value in this chapter. Then ask whether the second value could change while the first stays fixed. If it can, the two labels are not substitutes. For Visual Arts, this counterfactual is often clearer than a definition learned in isolation.
Swap one function column for a test-observer column and recompute without changing the spectra.
Evidence check: another reader should be able to reproduce the narrow result from the stored definitions, inputs and intermediate steps, then identify the nearest stronger claim that the evidence does not support.
A five-band table teaches structure and is not precision colorimetry.
CHAPTER 7 OF 20 · Work the mechanism
7. Separate instrument repeatability from observer spread
Repeated instrument readings estimate device and setup variability. Differences among people estimate inter-observer variability under the viewing protocol. Combining them hides which part of the evidence changed.
Trace the mechanism before interpreting the result. For Visual Arts, mark which value belongs to the source, which belongs to the measurement or layout algorithm, and which is a reported summary. A calculation is useful only when those roles remain attached.
Measure the pair three times and obtain nearly identical standard coordinates, while five observers give mixed match responses.
Practical transfer: build two records that share the headline number but differ in the condition discussed here. If the correct interpretation changes, the omitted condition is decision-relevant and belongs in the report. This test turns a vocabulary distinction into a reusable Visual Arts checking habit.
Introduce instrument drift but keep observers unchanged; the two uncertainty sources move differently.
Practical check: keep one counterexample beside the worked case. If the wording fails when that one condition changes, narrow the sentence until it describes only the evidence actually in hand.
Repeatability alone does not establish accuracy or perceptual agreement.
Observer metamerism occurs when a metameric match for a reference observer becomes a mismatch for a test observer with different responsivity. It is distinct from a change of illuminant.
Use a compact decision table with columns for condition, input, operation, result and limit. This makes the comparison reproducible and exposes where a missing field would force an assumption instead of an observation.
Keep the lamp and spectra fixed, then substitute a test-observer function in the calculation. Label the resulting colour difference as an observer-change result.
Misconception check: a plausible output does not prove the right reference, denominator, ancestor, spectrum or observation rule was used. Reconstruct one intermediate step and one changed-context prediction. If either cannot be stated, keep the result provisional rather than strengthening its wording.
Keep the observer fixed instead and change the illuminant; this is an illuminant-change question.
Transfer check: repeat the reasoning with one consequential condition changed while holding the rest fixed. State what changes, what stays comparable and why the original label alone was insufficient.
Real observers vary continuously; a small set of test observers is a model.
Retinal field size, fixation and surround affect which reference observer and adaptation assumptions are relevant. A phone patch and a gallery wall are not the same viewing task.
Preserve the raw inputs beside the derived result. If a later reader changes one condition, the original Visual Arts reasoning should still be recoverable without reverse-engineering a rounded headline or an unlabeled screenshot.
Compare a 2° central target with a 10° field using the same nominal chromaticity. Record angular size rather than only centimetres.
Evidence design: retain both a positive case and a near-miss. The positive case shows when the rule applies; the near-miss changes one consequential condition and shows when the conclusion fails. Together they teach the boundary of the Visual Arts mechanism better than another unlabelled example.
Move the viewer farther away so the physical patch subtends a smaller angle.
Audit check: retain the source version, units, settings, exclusions and rounding rule. A future correction should be able to alter one field without rebuilding the entire explanation from memory.
Field geometry is one factor and does not explain all individual differences.
Colour matches depend on illumination, background, adaptation time and luminance. Individual disagreement cannot be attributed to observer functions when the viewing conditions were not held reasonably stable.
Ask what would have to stay fixed for the comparison to be fair. Record those controls explicitly, then name the first uncontrolled condition that could reverse the conclusion. This is the practical transfer step, not an optional footnote.
Give viewers a stated adaptation period and neutral surround before the match task. Record deviations instead of silently excluding them.
Communication test: replace any sentence that says two things are “the same” or “different” with a sentence naming the operation, reference condition and supported consequence. The extra words are useful when they let a future reader repeat the decision without access to the original author.
Change the surround from neutral grey to saturated blue while keeping spectra fixed.
Decision check: end with an action or interpretation that follows from the stated conditions, followed by a boundary sentence explaining when a specialist rule, local procedure or additional measurement is needed.
A controlled booth still does not reproduce every real display or gallery condition.
CHAPTER 11 OF 20 · Audit the evidence
11. Keep age and physiology as limits, not labels
Lens, macular pigment and cone sensitivities can contribute to normal inter-observer variation. These mechanisms support a variability model; they do not justify inferring a participant’s biology from one mismatch.
Write the evidence in the order a reader can audit: the Visual Arts quantity, its reference object, the observation or browser rule, the transformation, and the decision being made. Keeping those fields visible prevents a familiar label from hiding a changed basis.
A model produces two plausible observer functions with different lens-density parameters. Compare their predicted match without assigning either to a named person.
First-principles test: ask what physical, observational or algorithmic operation produces each value in this chapter. Then ask whether the second value could change while the first stays fixed. If it can, the two labels are not substitutes. For Visual Arts, this counterfactual is often clearer than a definition learned in isolation.
Change one parameter at a time and identify which wavelength region drives the prediction.
Evidence check: another reader should be able to reproduce the narrow result from the stored definitions, inputs and intermediate steps, then identify the nearest stronger claim that the evidence does not support.
Individual health or vision claims require qualified assessment.
CHAPTER 12 OF 20 · Audit the evidence
12. Separate measurement uncertainty from model mismatch
Instrument calibration, spectral bandwidth and stray light create measurement uncertainty. A standard observer that does not represent a particular viewer creates model mismatch. They need different checks.
Trace the mechanism before interpreting the result. For Visual Arts, mark which value belongs to the source, which belongs to the measurement or layout algorithm, and which is a reported summary. A calculation is useful only when those roles remain attached.
A calibrated spectrometer repeats closely, yet a participant rejects the predicted match. Check observer-model limits before blaming random instrument noise.
Practical transfer: build two records that share the headline number but differ in the condition discussed here. If the correct interpretation changes, the omitted condition is decision-relevant and belongs in the report. This test turns a vocabulary distinction into a reusable Visual Arts checking habit.
Widen the instrument uncertainty interval and see whether it can explain the observed coordinate difference.
Practical check: keep one counterexample beside the worked case. If the wording fails when that one condition changes, narrow the sentence until it describes only the evidence actually in hand.
A disagreement can have several simultaneous causes.
Narrow spectral primaries can make observer differences more visible because small responsivity differences are sampled strongly. Standard coordinates remain useful but may not predict every cross-display match.
Use a compact decision table with columns for condition, input, operation, result and limit. This makes the comparison reproducible and exposes where a missing field would force an assumption instead of an observation.
Match a broad-spectrum display and a narrow-primary display for the standard observer, then collect paired viewer judgements under one protocol.
Misconception check: a plausible output does not prove the right reference, denominator, ancestor, spectrum or observation rule was used. Reconstruct one intermediate step and one changed-context prediction. If either cannot be stated, keep the result provisional rather than strengthening its wording.
Broaden the narrow primary while keeping the reference chromaticity close and predict reduced observer spread as a testable hypothesis.
Transfer check: repeat the reasoning with one consequential condition changed while holding the rest fixed. State what changes, what stays comparable and why the original label alone was insufficient.
Display-specific conclusions require measured spectra and sufficient observers.
Reflective samples combine illuminant and reflectance before observer functions are applied. A press proof and artwork can match instrumentally under one setup while some viewers report a mismatch.
Preserve the raw inputs beside the derived result. If a later reader changes one condition, the original Visual Arts reasoning should still be recoverable without reverse-engineering a rounded headline or an unlabeled screenshot.
Measure both reflectance spectra and compute under the reference illuminant and observer; then run a controlled paired viewing.
Evidence design: retain both a positive case and a near-miss. The positive case shows when the rule applies; the near-miss changes one consequential condition and shows when the conclusion fails. Together they teach the boundary of the Visual Arts mechanism better than another unlabelled example.
Change only the lamp to test illuminant metamerism separately from observer variability.
Audit check: retain the source version, units, settings, exclusions and rounding rule. A future correction should be able to alter one field without rebuilding the entire explanation from memory.
Surface texture, gloss and geometry can create additional appearance differences.
A useful visual experiment randomises order, controls viewing distance and surround, records uncertainty and repeats trials. It asks for a bounded match judgement rather than leading the viewer toward disagreement.
Ask what would have to stay fixed for the comparison to be fair. Record those controls explicitly, then name the first uncontrolled condition that could reverse the conclusion. This is the practical transfer step, not an optional footnote.
Present P–Q and Q–P orders across repeated trials and store response time plus match rating.
Communication test: replace any sentence that says two things are “the same” or “different” with a sentence naming the operation, reference condition and supported consequence. The extra words are useful when they let a future reader repeat the decision without access to the original author.
Remove labels and randomise left–right position to test cueing.
Decision check: end with an action or interpretation that follows from the stated conditions, followed by a boundary sentence explaining when a specialist rule, local procedure or additional measurement is needed.
Small classroom demonstrations illustrate logic and do not estimate population prevalence.
Transfer is shown when the mechanism predicts what happens after one controlled change. Changing observer functions, field size or illuminant should have different signatures.
Write the evidence in the order a reader can audit: the Visual Arts quantity, its reference object, the observation or browser rule, the transformation, and the decision being made. Keeping those fields visible prevents a familiar label from hiding a changed basis.
For one spectral pair, predict results for the standard 2°, standard 10° and one stipulated test observer before calculating.
First-principles test: ask what physical, observational or algorithmic operation produces each value in this chapter. Then ask whether the second value could change while the first stays fixed. If it can, the two labels are not substitutes. For Visual Arts, this counterfactual is often clearer than a definition learned in isolation.
Change the spectra to identical distributions; all observer models should then preserve the physical identity.
Evidence check: another reader should be able to reproduce the narrow result from the stored definitions, inputs and intermediate steps, then identify the nearest stronger claim that the evidence does not support.
Predictions depend on the quality of the spectral and observer models.
The sentence “the instrument says these colours are identical” overstates what was computed. The instrument and software report equality under a declared illuminant, geometry and observer convention.
Trace the mechanism before interpreting the result. For Visual Arts, mark which value belongs to the source, which belongs to the measurement or layout algorithm, and which is a reported summary. A calculation is useful only when those roles remain attached.
Rewrite it as “The pair matches within the stated tolerance under D65, 2° standard-observer colorimetry.”
Practical transfer: build two records that share the headline number but differ in the condition discussed here. If the correct interpretation changes, the omitted condition is decision-relevant and belongs in the report. This test turns a vocabulary distinction into a reusable Visual Arts checking habit.
Add “individual viewing was not tested” when no panel data exist.
Practical check: keep one counterexample beside the worked case. If the wording fails when that one condition changes, narrow the sentence until it describes only the evidence actually in hand.
Concise production language should remain tied to the actual decision.
A panel report names recruitment, screening, viewing conditions, trial count, response scale and distribution. One majority percentage does not erase consistent minority mismatches. Keep the trial-level record so a stable observer-specific mismatch is not mistaken for random response noise. Separate within-person repeatability from between-person spread, and state whether the task was an exact match, an acceptability judgement or a forced choice. A production panel can answer whether the tested pair is robust for that panel under the declared booth; it cannot turn the standard observer into a population average or establish how every future viewer will respond.
Use a compact decision table with columns for condition, input, operation, result and limit. This makes the comparison reproducible and exposes where a missing field would force an assumption instead of an observation.
Report three matches, one mismatch and one uncertain response, with repeat-trial consistency, instead of saying “80% correct”. Build a small matrix with observers as rows and repeated trials as columns, then preserve the uncertain category rather than forcing it into match or mismatch.
Misconception check: a plausible output does not prove the right reference, denominator, ancestor, spectrum or observation rule was used. Reconstruct one intermediate step and one changed-context prediction. If either cannot be stated, keep the result provisional rather than strengthening its wording.
Repeat on a second day and compare within-person and between-person variation. Change the response question from “same colour?” to “acceptable production match?” and explain why the decision threshold can change even when the spectra and observers do not.
Transfer check: repeat the reasoning with one consequential condition changed while holding the rest fixed. State what changes, what stays comparable and why the original label alone was insufficient.
A small panel supports the tested context, not universal human perception. Panel results also inherit the display, adaptation, ordering and instruction limits of the protocol.
A production note can combine standard colorimetry with a viewing panel while keeping their roles separate. The action threshold belongs to the project owner and consequences of mismatch. A museum proof, retail package and multi-display installation can rationally use different tolerances because the viewing duration, comparison opportunity and cost of failure differ. The report should therefore connect its evidence to one declared use, name any observer group that was actually tested, and preserve the spectral files so a later team can recompute under another standard observer or illuminant.
Preserve the raw inputs beside the derived result. If a later reader changes one condition, the original Visual Arts reasoning should still be recoverable without reverse-engineering a rounded headline or an unlabeled screenshot.
Approve the pair for a specified booth while requiring cross-display review for a narrow-primary installation. Record the instrumental tolerance and the panel acceptance rule as two fields rather than merging them into one pass mark.
Evidence design: retain both a positive case and a near-miss. The positive case shows when the rule applies; the near-miss changes one consequential condition and shows when the conclusion fails. Together they teach the boundary of the Visual Arts mechanism better than another unlabelled example.
Change the intended viewing population or display technology and revisit the decision. Keep the original approval as a bounded historical result instead of silently extending it.
Audit check: retain the source version, units, settings, exclusions and rounding rule. A future correction should be able to alter one field without rebuilding the entire explanation from memory.
Colour decisions can include aesthetic, accessibility and conservation factors beyond matching. Individual viewing evidence should never be used to diagnose a viewer.
Metamerism versus colour constancy, perceptual colour difference versus luminance contrast and simultaneous contrast versus physical change answer neighbouring questions. They should not be collapsed into one “colour perception” explanation.
Ask what would have to stay fixed for the comparison to be fair. Record those controls explicitly, then name the first uncontrolled condition that could reverse the conclusion. This is the practical transfer step, not an optional footnote.
Create a route table with trigger, measurement, observer condition and supported claim for each sibling.
Communication test: replace any sentence that says two things are “the same” or “different” with a sentence naming the operation, reference condition and supported consequence. The extra words are useful when they let a future reader repeat the decision without access to the original author.
Classify a new mismatch first by what changed: spectrum, lamp, observer, surround or luminance.
Decision check: end with an action or interpretation that follows from the stated conditions, followed by a boundary sentence explaining when a specialist rule, local procedure or additional measurement is needed.
Each sibling remains the owner of its own full mechanism.
Continue through the verified reading route
Metamerism and colour constancy
Metamerism and colour constancy. Use this when the changing condition is the light or the stability of one object.
Perceptual colour difference and luminance contrast
Perceptual colour difference and luminance contrast. Use this when two colour metrics answer different visual questions.
Simultaneous contrast and physical colour change
Simultaneous contrast and physical colour change. Use this when unchanged targets look different because their surrounds change.
Frequently asked questions
Does a standard-observer match mean everyone will see a match?
No. It is a reproducible model result under stated conditions, not a universal perceptual guarantee.
Is observer variation the same as instrument uncertainty?
No. They arise in different parts of the evidence chain and need different checks.
Should individual judgement replace colorimetry?
No. Production decisions can use both while keeping their roles and limits separate.
Authoritative sources and further reading
ISO/CIE 11664-1 standard colorimetric observers. The official standard specifies the CIE 1931 2° and CIE 1964 10° colour-matching functions and their viewing scopes.
CIE 080 observer metamerism. The official report defines mismatch caused by substituting a test observer for the reference observer.
CIE Colorimetry, 4th Edition. The official technical report covers standard observers, illuminants, tristimulus calculations, colour differences and advanced colorimetry.
Return to the Visual Arts subject guide, The Gold Standard of Education, or the How X Works hub.