The e(M/P)ty promise (MAGAZINE)

June 10, 2020
Researcher MARK REA at the LRC warns against using the M/P ratio as a circadian photometric method.

In response to recent LEDs Magazine State of the Industry survey coverage and commentary on increasing interest in applications of lighting for health and wellbeing, well-known lighting researcher Mark Rea has raised concerns about a photometric measure gaining attention in the field. Known as the melanopic/photopic ratio, it is being used to characterize light sources designed for stimulating circadian response in humans. For solid-state lighting (SSL) product developers and even lighting specifiers, it is important to consider all the parameters and limitations of circadian metrics so as not to release commercial products or implement lighting designs that could set back human-centric or healthy lighting efforts. — CARRIE MEADOWS

Understanding S/P ratio

The scotopic-photopic (S/P) ratio has been a useful way to describe the spectral characteristics of visual stimuli that need to be seen in the periphery, such as deer entering the roadway.1,2 Peripheral detection depends exclusively upon rods [scotopic, V’(λ), spectral sensitivity] under starlight and exclusively upon cones [photopic, V(λ) or V10(λ), spectral sensitivity] during the day.

The S/P ratio has been particularly useful for characterizing street lights, because at street lighting levels both rods and cones are operating simultaneously. Since V’(λ) represents the spectral sensitivity at very low light levels and V10(λ) at high light levels, any street light, no matter its spectral power distribution, can be described simply in terms of its S/P ratio. For physiological reasons, then, the detailed spectral power distributions (SPDs) of street lights can be legitimately simplified in terms of their S/P ratio for peripheral detection.

Perhaps it is obvious that the S/P ratio cannot be used alone to predict off-axis detection. One must also know the operating range, or adaptation level, of the visual system. So depending upon whether the ambient lighting is bright, dark, or dim, the relative effectiveness of light sources with different S/P ratios will vary. Those sources with a high S/P ratio will be most effective under a given, low photopic light level, while those with a low S/P ratio will be most effective under a given, high photopic light level. In other words, light level is a critical determinant for selecting the most effective street light. Perhaps less obvious, the S/P ratio is not helpful for characterizing the effectiveness of low-beam headlights that are designed to illuminate the roadway. For central foveal vision used to navigate the roadway, only cones are operating, so only photopic conditions apply.3 Thus, the S/P ratio is only useful for describing the spectral characteristics of light sources where the lighting design objective is to detect objects entering the road at street lighting levels.

M/P as a circadian metric

It has been recently argued that, analogous to the S/P ratio, the melanopic-photopic (M/P) ratio can be “appropriate” for characterizing light sources intended for circadian system stimulation.4 Having a system of photometry that can characterize the effectiveness of a light source for stimulating the human circadian system better than one based upon V(λ) is quite important. In this context, good lighting practice would provide energy-efficient lighting for activating the circadian system during the day and for minimizing activation of the circadian system during the night. The lack of a robust, 24-hour cycle of biologically-activating light during the day and inactivating light during the night could lead to sleepiness during the day, poor sleep at night, and depression; it could even contribute to more serious maladies like heart disease and breast cancer.

The lynchpin between the retina and the biological clock is the intrinsically photosensitive retinal ganglion cell (ipRGC) that contains the photopigment melanopsin.
Superficially then, the M/P ratio is attractive because it appears to have parity with the S/P ratio as a useful metric for characterizing the effectiveness of different light sources for activating (or inactivating) the circadian system at a given photopic light level. Unlike the S/P ratio — which does have some utility for street lighting — the M/P ratio has no value for any lighting application.

The figure below illustrates how poorly P [based on V(λ)] and M [the spectral sensitivity of the photopigment melanopsin] characterize the spectral sensitivity of the circadian system to narrowband light stimuli. The same is true for polychromatic (white) light sources5 because, unlike the S/P ratio that is rooted in an understanding of the physiology of peripheral vision, the M/P ratio has no physiological significance for circadian activation. Therefore, the M/P ratio would not accurately represent the relative effectiveness of different light sources for stimulating the circadian system at the same photopic light level. Of course, the absolute amount of light must also be known to accurately assess the effectiveness of different light sources. So, like the S/P ratio, the M/P ratio is inherently an incomplete description of the effectiveness of a lighting system for stimulating the circadian system.

What, then, would be the purpose of the M/P ratio? Since the M/P ratio has no physiological foundation, the only imaginable justification for its use in lighting might be to transform a photopic unit, like lux or candelas, into a term that would include short wavelengths and thereby more closely represent the spectral sensitivity of the circadian system. This photometric half-measure would not only be inaccurate because there is no physiological support for the M/P ratio, it would be perniciously misleading because the reference to the photopigment melanopsin implies such support. For the unsophisticated regulator or even many lighting specifiers, the M/P ratio suggests a scientific foundation, but is, in fact, scientifically unjustified.

Alternative metrics in play

There are two alternatives to consider with the aim of having a system of photometry that can better represent how light affects the circadian system. The first alternative would be based upon a metric like circadian stimulus (CS) that has a physiological foundation as well as predictive utility for circadian system response. In fact, enough was known to move forward with this approach for a lighting design guideline aimed at improving circadian regulation (UL DG 24480)6 recognizing that as science progresses, refinement to that understanding can and should lead to revisions of the metric.

The second alternative would be based upon a metric unrelated to the complexities of neurophysiology, but would simply expand the spectral range of the orthodox photopic luminous efficiency function [V(λ)] used in all commercial photometry to include human retinal sensitivity to all visible wavelengths — including those short wavelengths that stimulate the human circadian system. V(λ) is based upon the spectral sensitivities of only two of the five known photoreceptors in the retina. It is spectrally too narrow to properly represent light for the human visual and non-visual systems.

To address this problem, a universal luminous efficiency function, U(λ), was developed from all five retinal photoreceptors7, thereby eliminating the inherent bias of V(λ) against short wavelengths that stimulate the human circadian system, which, supposedly, was the point of the M/P ratio. By simply broadening the spectral envelope to fully represent optical radiation as it stimulates the human retina, visual as well as non-visual system sensitivities are not inadvertently penalized.

Importantly, light source selection for healthy lighting can be supported without the pretense that the complexities of circadian phototransduction have been characterized. So, either alternative — one based upon the non-linear complexities of the retina (CS) or one based upon the full spectral sensitivity range of photoreceptors in the retina [U(λ)] — would not obfuscate, as V(λ) and the M/P ratio do, the effectiveness of different light sources as they can stimulate the human circadian system. Most crucial is that specifiers interested in promoting healthy lighting would not be encouraged to believe an “e(M/P)ty” promise.


1. T. Goodman et al., “Mesopic visual efficiency IV: a model with relevance to nighttime driving and other applications,” Light Res. Technol., 39, 365–392 (2007).

2. M.S. Rea, J.D. Bullough, J.P. Freyssinier-Nova, and A. Bierman, “A proposed unified system of photometry,” Light Res. Technol., 36, 85 (2004).

3. Y. He, A. Bierman, and M.S. Rea, “A system of mesopic photometry,” Light Res. Technol., 30, 175–181 (1998).

4. N.J. Miller and A. Irvin, “M/P ratios – Can we agree on how to calculate them?LD+A (Feb. 2020).

5. R. Nagare, M.S. Rea, B. Plitnick, and M.G. Figueiro, “Effect of White Light Devoid of ‘Cyan’ Spectrum Radiation on Nighttime Melatonin Suppression Over a 1-h Exposure Duration,” J. Biol. Rhythms, 34, 195–204 (2019).

6. Underwriters Laboratories Inc. Design Guideline for Promoting Circadian Entrainment with Light for Day-Active People, Design Guideline DG 24480, Edition 1 (2019).

7. M.S. Rea, “The lumen seen in a new light: Making distinctions between light, lighting and neuroscience,” Light Res, Technol., 47, 259–280 (2015).

8. G.C. Brainard et al., “Action spectrum for melatonin regulation in humans: evidence for a novel circadian photoreceptor,” J. Neurosci., 21, 6405–6412 (2001).

9. K. Thapan, J. Arendt, and D.J. Skene, “An action spectrum for melatonin suppression: evidence for a novel non-rod, non-cone photoreceptor system in humans,” J. Physiol., 5 35, 261–267 (2001).

Get to know our expert

MARK REA, PHD, is professor of architecture and cognitive sciences at the Lighting Research Center (LRC) at Rensselaer Polytechnic Institute. He served as LRC director from 1988 to 2017. Rea is well-known for his research in circadian photobiology, mesopic vision, psychological responses to light, lighting engineering, and visual performance.

For up-to-the-minute LED and SSL updates, why not follow us on Twitter? You’ll find curated content and commentary, as well as information on industry events, webcasts, and surveys on our LinkedIn Company Page and our Facebook page.