Preference for the Color Rendition of Skin

Quite a bit is known about group preference for how light sources render object colors, as distilled into Annexes E and F of TM-30-20 (ANSI/IES 2020). One area that is weakly substantiated, however, is group and subgroup preferences for how light sources render skin. This is a conspicuous shortcoming because people are intended to be the primary beneficiaries of lighting.

Preference for the color rendition of skin is a challenging scientific problem with cultural implications. Skin spectral reflectance functions (SRFs) are generally reported as averages over a small area of skin, usually measured on some part of the arm or hand. While people with darker or lighter skin have variation in mean reflectance, the spectral features do not vary as much as one might expect (e.g., Angelopoulou 2001; Cooksey et al. 2019). The presumed implication is that it may be possible to predict reasonably well the color shift of a range of skin SRFs with a small number of samples.

With skin, however, average shifts should not be expected to tell the whole story. If the color of a toy or piece of clothing is altered by a light source, it is not personal. But if a light source distorts a person’s skin in a way that is undesirable, then that is intrinsically personal.

Evaluating skin rendition is further complicated because skin is not homogenous and cannot be represented in a single SRF, skin tone is dynamic based on blood flow and other factors, and preference varies with ethnicity, may vary with race, may be informed by culture, and may vary with individual characteristics. For example, some people may prefer the ruddier skin appearance associated with red-chroma-enhancing light sources, while others may not (e.g., those with neutral or cool undertones, mottled skin, or a skin condition). There is some evidence for cultural differences—e.g., some evidence suggests that, on average, Caucasians prefer a color-shift that causes a ruddier complexion more so than Asians (Wei et al. 2014).

TM-30-20 does not include a specific design intent or design guidance related to skin appearance. This not because there is lack of interest, but because there is not enough knowledge to produce defensible recommendations. “More research is needed” is a common cry among researchers, but in this case the cry is true. It will be inspiring if future work includes an ethnically, racially, and culturally diverse research team that contributes to the conceptualization and design of new experiments, and a diverse set of participants that permits analyses based on ethnicity, race, and culture. This sort of granular information would provide deeper understanding of how different groups of people experience light and may open opportunities for spectrally tailored lighting solutions.