Short answer
Colour-matching functions describe an average observer. Real eyes differ in lens yellowing, macular pigment and the exact sensitivity of their cones, so a pair that matches for one person — or for the CIE standard observer — can mismatch for another under the very same light. This is observer metamerism, and it grows as spectra get narrower.
Colorimetry works because the CIE defined a standard observer: colour-matching functions averaged from experiments on a small number of people with normal colour vision. The 1931 observer describes a 2° field, roughly a thumbnail at arm's length; the 1964 observer describes a 10° field, closer to how rooms and large samples are seen, and is the one TM-30 uses. Instruments compute colour for this average eye. No individual has exactly these functions, and the CIE has long recognised that a pair matched for the standard observer is not guaranteed to match for everyone, publishing a special metamerism index for a change in observer in 1989.
Three differences dominate. The crystalline lens absorbs short wavelengths and yellows with age, so an older eye receives less blue and violet. Macular pigment, concentrated in the centre of the retina, absorbs blue and varies several-fold between people, which is also why the 2° and 10° observers differ. And the cone photopigments themselves vary slightly in peak wavelength. The CIE's physiologically based observer, CIE 170-1:2006, builds colour-matching functions from cone fundamentals and allows the field size and the age-dependent lens absorption to be adjusted, precisely so that such differences can be modelled rather than ignored.
When two stimuli have broad, smooth spectra, small shifts in an observer's sensitivities change both similarly, and the match tends to survive. When one or both are narrow-band — laser projectors, quantum-dot or OLED displays, RGB LEDs — the signal depends on exactly where a narrow peak falls on each observer's sensitivity curves, and small individual differences produce visible disagreement. This is why two people can see a calibrated wide-gamut monitor and a printed proof differently even when both are 'matched' by instrument, and why display calibration for colour-critical work increasingly considers observer variation. The site's digital colour section covers the screen side of this.
Disagreement over a near-match is not necessarily one person being wrong. First rule out the physical causes — different light, angle, surround, adaptation, or a colour vision deficiency, which is a larger and different effect. If the disagreement persists under controlled conditions, observer metamerism is likely, and the answer is to reduce spectral difference between the two items rather than argue: same colourants, or a spectral rather than visual match. Where one person must approve colour, keep the same person and conditions from start to finish.
Why: Observer metamerism between the proof's pigments and the screen's narrow primaries.
Fix: Compare print to print under a graded daylight simulator; do not arbitrate a print by a screen.
Why: Lens yellowing reduces short-wavelength input, shifting matches involving blues and violets.
Fix: Keep one approver, or match spectrally so the pair agrees for everyone.
Each statement is labelled by kind — established fact, a standard’s requirement, observed market data, a convention, or Colourwise’s own interpretation or analysis — with the strength of the evidence behind it.
StandardStrong evidence
The CIE's special metamerism index for a change in observer evaluates how a match made for the reference observer breaks down when a test observer with normal colour vision is substituted.
Source: CIE 080-1989 Special Metamerism Index: Change in Observer
StandardStrong evidence
CIE 170-1:2006 defines cone fundamentals and colour-matching functions for field sizes from 1° to 10° and allows adjustment for age-related lens absorption.
Source: CIE 170-1:2006 Fundamental Chromaticity Diagram with Physiological Axes – Part 1
FactStrong evidence
TM-30 calculations use the CIE 1964 10° colour-matching functions, and swapping in a different set tends to have a small effect on the results.
Colourwise interpretationModerate evidence
Observer metamerism is more consequential for narrow-band stimuli such as laser and quantum-dot displays than for broad-spectrum surfaces under daylight.
Based on: Follows from the colorimetric integral: with a narrow peak, the result depends on the local value of each observer's sensitivity rather than an average over a wide band.
Reviewed 29 September 2026. Colourwise summarises its sources in its own words and does not reproduce standards text or proprietary colour data. Spotted an error? Tell us.