The human visual system: why no two assessors see the same thing

Two experienced colour assessors look at the same textile sample under identical lighting. One approves it, the other hesitates. No difference in training, no difference in procedure — yet a difference in perception. How can that be?

The answer lies in the instrument we use to perceive colour: the human visual system. Unlike a spectrophotometer, that instrument is not standardised. It differs from person to person, changes with age, and has built-in limitations that few professionals fully recognise.

This article explains how the visual system processes colour, where the variations lie, and what that means for anyone who assesses colour professionally.

Three sensors, one colour image

All colour perception begins in the retina, in cells known as cones. The human eye contains three types, each sensitive to a different part of the light spectrum:

• S-cones — peak sensitivity around 420 nm, the blue range.
• M-cones — peak sensitivity around 535 nm, the green range.
• L-cones — peak sensitivity around 565 nm, the red-orange range.

Note that the peaks of the M- and L-cones lie close together — only 30 nm apart. This makes biological sense: it enables us to perceive subtle colour differences in the yellow-green-orange region, precisely the range that was essential for our ancestors in distinguishing ripe fruit from unripe fruit.

The cones are concentrated in the fovea, an area only 1.5 mm in diameter at the centre of the retina. The fovea contains only cones — no rods — and is responsible for the sharpest colour perception. Everything you look at directly is processed here.

Outside this area, cone density drops rapidly and rods take over. Rods are extremely sensitive to light but colour-blind. This is why your peripheral field of view registers movement and brightness, but hardly any colour.

From light to colour experience: five steps

The path from a ray of light to a conscious colour experience is shorter than you might think — but more complex than it appears. It unfolds in five steps.

Step 1 — Phototransduction. Light reaches the retina and is absorbed by the photopigments in the cones. This triggers a biochemical cascade that converts the light signal into an electrical signal. Each type of cone responds to its own wavelength range.

Step 2 — Neural processing in the retina. The electrical signals are processed directly in the retina by bipolar cells and ganglion cells. Already at this stage, colour coding begins according to the opponent process theory: signals are converted into red-green, blue-yellow and light-dark channels. The retina is therefore not a passive sensor — it is an active pre-processor.