Toward adaptation-aware optimization of tunable multi-LED lighting: modeling illuminance-dependent hue-bin preference targets

Adaptive spectral tuning of solid-state lighting requires perceptually grounded models of color preference that account for adaptation. This study quantifies how adaptation illuminance at three levels modulates preferred local chroma and hue shifts in four TM-30 hue bins (1: red, 4: yellow, 6: green, 12: blue) using ecologically valid, multi-object stimuli and a dual-lighting design that allowed independent modulation of object color appearance and adaptation illumination. Pooled observer ratings from 51 participants were modeled with bin-wise bivariate Gaussian functions in local chroma-hue shift space, capturing both location and anisotropy of preference distributions. Preference maxima were estimated via bootstrapping, and model contours defined confidence ellipses. Results show a consistent logarithmic decrease in preferred chroma shift with increasing illuminance, while hue-shift trends vary by bin, accompanied by decreasing tolerance regions at higher illuminance. Compared to former studies, the findings predict preference for higher chroma levels under typical indoor scenarios and exhibit a less steep decline with increasing illuminance. The present results provide broader, adaptation-aware, bin-wise mappings suitable for predictive preference metrics and adaptive lighting optimization.