View Full Version : What spectrum of light is the human eye most sensitive to?

04/05/2006, 01:27 AM
I remember a couple professors telling me that the human eye is most sensitive to blue light. I have heard the argument that the human eye is most sensitive to red light as well and I objected...but from a different POV...this could be considered true.

I see in your Reefkeeping segment, that you say 550nm/green is what human eyes are most sensitive to. Hmmmm...yet another curveball.

Now, Im not trying to argue here...Im civil, you are the Professor, but it seems that this subject has come up for debate a few times depending on who you talk to...so I was hoping to have a discussion with you Sanjay, to gain a better understanding if nothing else.

The argument I have heard for red light being the spectrum our eyes are most sensitive to is from professional photographers...so there might be a lack of radiometric measure for exact verification...but their opinion is based on when they fire up a 1000watt tungsten filament bulb, this is the brightest bulb they know of. For some daylight settings, they use bluer 1000watt halide bulbs, but they say that the halide is no where near as intense to look at. This is kind of odd since the PAR of a tungsten should pale in comparison to that of a halide, right? Yet the tungsten looks brighter...hmmm.

The argument for blue light being the one our eyes are most sensitive to is based on a few things. First, nightvision. Blue light is used by BMW to provide the best supplimental light with those xenon headlights...a frequency that we are most sensitive in low levels, yet of a spectrum that doesnt disrupt our nightvision. How are we most sensitive? Well, unlike other frequencies, the concentration of cones in the fovea centralis that actually pick up blue light are only 2% (and 32% green sensitive cones, and 64% red cones). This is because blue light is part of our perception of what daylight is...and our rods actually pick up blue frequencies rather than our cones. So the argument can be made that since there are only 2% of our cones that pick up blue light, these must be the most 'sensitive', or we would hardly be able to see blue light in comparison to all the other spectrums. And since blue light is linked to our perception of brightness more than with red or green frequencies, it could be said that blue light is what we are most sensitive to. Of course, the fact that 64% of our cones are attuned to red light suggests that red light is what we are most sensitive to...so what I am trying to get at is that it alll depends on how you interpret the data. Lets say you have a color blind person...their lack of cones would mean that the only color they see and interpret as brightness would be blue light, as rods pick up blue. This is also why blue light is so great for nightvision...it allows us to see by giving us a 'daylight' or sorts without triggering our cones. So I suppose the bottom line to this argument is that since rods see blue light, and no other frequencies, as daylight, we would be most sensitive to it since not only do our cones 'see' it, but so do our rods. Blue light has more to do with what we consider 'brightness' than other frequencies.

But this might only be true of lower intensity situations. As a lights intensity increases, it appears that our rods will take the back burner to the cones and the cones interpret the color...and then its all up for grabs. But blue light could be argued to be the color we are most sensitive to.

Here are some of my sources...



There was also the evolution based argument, based not on the eye, but how our brain interprets the data. It suggests that red is the most 'vivid' since in our evolution red means blood...pain, injury, fighting, warning, or food. Blue is less 'alarming' to us since it represents sky and maybe water...two things of little consequence in our daily survival. Green would also be the most common...leaves and trees would be green all over and so would the grasses...also of little inportance. Of course, this is a more having to do with perception than what we see...

Whats your take on the info? I still dont quite see how you came up with Green as the most sensitive as far as our eyes see... I can see how arguments for red or blue originate depending on the brightness of the light...but green?

04/05/2006, 08:47 AM

What I was trying to say is that the collective response of the human eye, has a peak at 555 nm (photopic vision) and a peak of 507 (scotopic vision) as established by experimental data on human subjects. The response of the human eye to various wavelengths peaks at 555nm, so at equal intensities green looks brighter than the red and blue.

Using the same site that you refer to,


“The Photopic efficacy curve was extrapolated from testing done on 'Standard Observers'. This was done by taking a person with normal vision, and having them compare the brightness of monochromatic light at 555 nm, where the eye is most sensitive, with the brightness of another monochromatic source of differing wavelength. To achieve a balance, the brightness of the 555 nm source was reduced until the observer felt that the two sources were equal in brightness. The fraction by which the 555 nm source is reduced measures the observer's sensitivity to the second wavelength. This exercise is repeated through many wavelengths and many observers. The average of the results gives us the relative sensitivity of the eye at various wavelengths. In 1924, the International Commission on Illumination adopted the "relative sensitivity curve for the C.I.E. Standard Observer".

I guess the word "sensitive" is sensitive to how its used :-)


04/05/2006, 01:17 PM
Thanks Sanjay. It really is a 'sensitive' subject, depending on the the situation I suppose...lol.

04/26/2006, 03:53 PM
Sanjay, I just read part III of your Facts of Light article. It is a great article, reflective (ahem) of points that were in tremendous debate in the reefkeeping hobby seven or so years ago. I'm glad that you could put the weight of science behind it in such simple language.

In response to Hahnmeister's comments, the sources of "sensitivity" he cites are indeed not related to the physical element of light sensors detecting (or responding to) irradiant energy. Most of the effects he's talking about are related to psycho-visual effects, that is, the way the brain interprets the signal it gets from the nerves in the retina.

In terms of the actual sensitivities, remember that the different cones are responsive in different amounts to different ranges of the spectrum, and that rods respond nearly uniformly to all parts of the spectrum. The "green" cone produces the strongest response of all the cones in the 550nm range.

Some of the psycho-visual elements cause the pupil to dilate or contract, which has far more to do with why BMW chooses blue light for night driving than the sensitivity of the rods in the eye to light energy, and why a red (or yellow) light "looks" brighter than a blue one. The rods also carry a substantially reduced ability to resolve detail, so the argument about using blue headlights to stimulate the rods doesn't really hold up. Blue light does allow smaller details to be resolved, but the blue cone produces a weak response compared to the Red and Green ones. This accounts for why objects lit with blue light appear "crisper".

A better example of green"sensitivity" would be the 6500K metal halide light bulbs themselves. Your own tests some years ago showed strong peaks in the green range. These peaks are there by design, to maximize the usable light energy for people performing tasks under these lights. The lighting industry has been designing light bulbs this way for years. This is precisely why using a lux meter to approximate PAR is bad, because it emphasizes the green region of the spectrum for human sensitivity, but that region doesn't correspond well to the regions of the spectrum that various chlorophyls respond to most strongly, so the 6500 bulb has a high lux reading, but that may or may not correspond to a strong PAR reading, or (even better IMHO) PUR reading.

04/26/2006, 11:50 PM
Thanks for the insight GreenManMaple. And welcome to RC!!! (see you have post #1 there).

I seem to remember something about resolution varying with the spectrums as well...blue light making it harder to see detail and red being easier...or the other way around. It was in the links I cited.

So, for the most part... Green really is the spectrum we are most sensitive to...hmm.