Unique hues as revealed by unique-hue selecting versus partial hue-matching

Abstract

Unique hues are usually defined as those that cannot be introspectively reduced to any other hue. According to a major dogma of color science, there are four unique hues: yellow, blue, red, and green. Yet only 55 of the 173 inexperienced observers who participated in our experiment selected exactly four Munsell papers that, according to their judgment, had a unique hue. The number of papers selected by the rest of the observers varied from zero to nine. We believe that such variability of unique hue selection is due to the ambiguity of the introspective criteria for hue uniqueness. Along with the traditional technique of unique hue selection, an alternative method based on partial hue-matching has also been used to establish the nomenclature of unique hues. The partial hue-matching method is based on observer judgments concerning the presence of a common hue in a pair of colors. Observers are not supposed to name (or make any other judgments of) this common hue. Without presupposing their number, the unique hues are derived from the observer's responses to a sample of color pairs. The results obtained by this new method generally support the classical notion of four unique hues.