Colour preference of lighting is generally influenced by three kinds of contextual factors, which are the features of light, object and observer. In this study, a series of psychophysical experiments were conducted to investigate and compare the effect of certain factors on colour preference, including spectral power distribution of light, lighting application, observers' personal colour preference, regional cultural difference and gender difference. Certain LED lights with different correlated colour temperatures were used to illuminate a wide selection of objects. Participant response was quantified by a 7-point rating method or a 5-level ranking method. It was found that the preferred illumination for different objects exhibited a similar trend and the influence of light was significantly stronger than that of other factors. Therefore, we conclude that the light itself (rather than, for example, the objects that are viewed) is the most crucial factor for predicting which light, among several candidates with different CCTs, an observer will prefer. In addition, some of the gamut-based colour quality metrics correlated well with the participants' response, which corroborates the viewpoint that colour preference is strongly influenced by colour saturation. Moreover, the familiarity of the object affects the ratings for each experiment while the colour of the objects also influences colour preference.

No abstract

In this work, the correlation between the whiteness of lighting and the preference for the colour of lighting was comprehensively investigated based on four psychophysical studies. Different sets of experimental light sources, two of constant correlated colour temperatures (CCT) and two of different CCTs were adopted to illuminate multiple experimental objects with different illuminance levels (200 lx or 500 lx). The observers were asked to rate their colour preference towards the illuminated objects and rate their white perception towards the illuminated environment of the empty light booth. It was found that for the studies with same CCT but different chromaticities, people indeed preferred whiter illumination. However, for the cases with multiple CCTs, that statement is only valid between 2500 K and 5500 K, since too white a light source impaired colour preference. In addition, based on the whiteness rating data, the performance of three typical metrics for white lighting was assessed. The findings of this study may contribute to a deeper understanding of colour and whiteness perception in lighting, especially for lighting designers and manufacturers.