Abstract
The locus of unique blues, white, and unique yellow divides colour space into two colour categories: those with reddish hues and those with greenish hues. We have shown that colour discrimination is optimal at this fundamental boundary under adaptation to a metamer of D65 (Danilova, Mollon, 2012, Vision Research, 56, 162). We suggested that both the phenomenological boundary and the locus of optimal discrimination correspond to the equilibrium state of a channel that draws signals of one sign from long- and short-wave cones and of opposite sign from middle-wave cones. The present experiment asked whether the phenomenological equilibrium and the locus of optimal discrimination would be concomitantly shifted by chromatic induction. Our inducing fields resembled those of Monnier and Shevell (2004, Vision Research, 44, 849). A target annulus was embedded within concentric rings of alternating chromaticities. The eye was adapted to D65. The inducing fields were of the same brief duration as the target (150 ms) but were dimmer: our reasoning was that colour induction depends on ratio signals and thus dim inducers may be as effective as bright ones, while casting less stray light into the target annulus. We measured 2AFC discrimination thresholds along lines orthogonal to the red-green boundary. In interleaved runs, the subjective hue boundary was measured empirically for each condition and for each observer. The phenomenological and the performance measures were similarly displaced. We propose that both depend on the same neural channel and that chromatic induction shifts the equilibrium point of this channel.