|The experiment called on subjects to compare the "naturalness" and "vividness" of colors under different lighting conditions.
Color rendering index (CRI) is the most common metric used by the lighting industry to represent the color rendering properties of electric light sources. CRI was intended to characterize how “true” or “natural” objects appear when illuminated by a light source. However, it was never intended to characterize other aspects of color rendering for different applications, including how well object colors can be differentiated or how saturated colors appear. No single metric can capture all the dimensions of color rendering. Current recommendations for neonatal intensive care units (NICU), for example, call for the use of combined color rendering metrics.
LRC researchers examined CRI and two other metrics of color rendering to determine how useful they are at predicting subjective judgments of how “natural” objects appear, how “vivid” objects appear, and how well one can discriminate between subtle differences in hue. The U.S. Environmental Protection Agency (EPA) sponsored the project.
Researchers performed three experiments using different phosphor-based light sources. One experiment used only low (<4400) CCT sources; another used only high (>5000) CCT sources; and the third used both.
Each experiment used two types of tasks under four different light sources and under two light levels. The color discrimination task required subjects to perform the Farnsworth-Munsell 100-hue test. A paired-comparison task presented subjects with a collage of photos showing two species of birds (blue jays and red cardinals). Subjects compared the “naturalness” and “vividness” of the blue and the red plumage under the different lighting conditions. They also assessed the overall “naturalness” and “vividness” of the collage itself.
When used in conjunction with another measure of color rendering (gamut area index [GAI]), CRI is useful at predicting subjective judgments of how “natural” and “vivid” objects appear, and how well one can discriminate between subtle differences in colors.
“Neither measure by itself meets all of the expectations of a light source for providing good color rendering. Both are needed to ensure good color rendering,” explained Mark Rea, the director of the LRC and principal investigator of this project. Light level is also important for color rendering, Rea added.
Rea says we need more research to determine if just two metrics are sufficient to ensure good color rendering from a light source and whether CRI and GAI are the best metrics to use. In the meantime, he says it is clear that using these two-metrics together will be a useful improvement over relying solely on CRI as the measure of color rendering.
The scientific paper, “Color rendering: A tale of two metrics,” will appear in the journal, Color Research and Application.