Perceiving white light from Lightbulb Illumination

[mk9898]

What is exactly hitting the retina for us to perceive white light in a room radiated by light from a lightbulb?

The light from the lightbulb caused by heat radiates wavelengths randomly through the space around it. But how could it be that the superposition of these random frequencies and wavelengths end up being close to white?

 

[brainpushups]

Color is due to perception, it is not a physical quality. The the trichromatic theory of color vision would be a good starting place to read, but this theory cannot explain many aspects of perceived colors (e.g. colored shadows). The perceived color of an object is due to the colors that surround the object as has been well-documented by experiments initially performed by Edwin Land. He came up with a theory of 'color constancy' which would be another good thing to read up on.

Edit: by 'physical quality' I mean that the wavelength intensities that enter the eye cannot alone account for the color that is perceived (again, as demonstrated by Land's experiments).

 

[mk9898]

Thanks I didn't know there were still theories on this. I tried to take the geometrical description of light rays and tried to use it to describe what exactly is hitting the retina for me to observe light but then the whole superposition of the different, incoherent light waves made it tough to really imagine that there is just a constant beam of light shooting every which way but at the same time appearing continuous to the eye. Run on sentence but it's a good intro into learning German ;).

 

[brainpushups]

A quick google search just turned up this, which I haven't read yet, but it describes one of the most interesting color experiments that Land performed (it seems, with my quick skim, that it was an accident!) in which a full color photograph can be obtained by overlaying photos taken through red and green filters (one each) and then imposing a red filter between a projector and screen. The trichromatic theory would predict an image of red hues (due to white/red light combinations) but the photo is full color!

There is a nice BBC documentary about the color constancy theory that came out back in the 70s I think.

 

[Drakkith]

There's one very important thing you always have to remember when dealing with color. Color is subjective. The colors you see exist because your mind and visual system do a great amount of processing the signals from your cone cells in your retinas. They can even compensate for changes in the overall color of the ambient light. That's why you can still identify 'white' paper under a yellowish light coming from a lamp in your house. A perfect example is the fact that my Dad is colorblind and does not see the same colors that I (and most others) do. For him, green streetlights and white parking lot lights are identical in color.

The entire process is quite complicated. I'd start with understanding how the cone cells in your retina work and then look into how the brain and visual system process this information. A few links:

https://en.wikipedia.org/wiki/Color_vision
https://en.wikipedia.org/wiki/Visual_perception
https://en.wikipedia.org/wiki/Visual_system

 

[A.T.]

...incoherent light waves made it tough to really imagine that there is just a constant beam of light shooting every which way but at the same time appearing continuous to the eye...

Look at the frequencies of visible light. Neither the sensory cells in the retina, nor the nerve connections to the brain, nor the brain itself are fast enough to capture the actual wave from.

Instead you have different sensory cells with shifted frequency dependent sensitives. When all types get stimulated "equally" you see white. But "equally" is adaptive, as you brain might perform some color balance.
https://en.wikipedia.org/wiki/Cone_cell

 

[Eric Bretschneider]

You would be better off referring to the wavelengths of light rather than the frequencies. Yes, I know you can calculate one from the other, but photoreceptor action spectra, photopic response and tristimulus response curves are all given in terms of wavelength (nm).

As you alluded, the visual system in the brain acts in a manner similar to white balance for cameras. What we perceive depends on what we have recently seen and the entirety of the visual field.

 

[A.T.]

You would be better off referring to the wavelengths of light rather than the frequencies.

Not for the point I was making, about the impossibility to capture the actual incoherent EM-waveform, that the OP seems to imagine as the sensory input.

 

[Eric Bretschneider]

Since virtually all natural light is incoherent, it wouldn't make much sense to respond to comments of incoherent vs coherent. We are talking about biological systems, not advanced electronics.

The fastest neurons can respond at up to about 3 kHz.

I was also trying to point the OP to information that is easier to find.

 

[DaveC426913]

... this theory cannot explain many aspects of perceived colors (e.g. colored shadows).

At the risk of derailing, why not?

If I have very low level ambient light, and a red spot light with an obstruction, the wider area will appear red. The shadowed area will appear not quite black, so it will be very dim white, minus red = cyan. i.e. a cyan shadow.

 

[brainpushups]

At the risk of derailing, why not?

I'm not quite sure what you are asking. My point was that a colored shadow is not due to to actual wavelengths entering the eye. My understanding is that the perceived colors are due to the combined response of the different cone cells (most people have three types) and the intensity of the three wave bands (long, middle, short) stimulating them. I've read that Land demonstrated full color vision is possible with just two narrow wavelength bands. Wikipedia says the two wavelengths used in his experiments were 579 and 599 nm. There is a cited Scientific American article to go with that information, but I have not read it. Here's a link to it: http://www.psy.vanderbilt.edu/courses/psy236/ColorVision/Land1959.pdf

 

[haruspex]

how could it be that the superposition of these random frequencies and wavelengths end up being close to white?

First, you need to define white. You see a lot of things around that you would say are white, but placed next to each other you find some to be whiter than others. Ok, that still suggests there is a perfect white, effectively the whitest thing you ever see. (Could there be A, B and C such that A looks whiter than B looks whiter than C looks whiter than A? I doubt it.)
Colour perception is a response to the signal levels emitted by your three types of cone receptor and your grey-scale rod receptors, and, as others have noted, is relative to the light from the rest of the view.

What we perceive as perfect white seems to match the spectrum of light from the sun. Yes, we see a yellow sun and a blue sky because the wavelengths get separated in the atmosphere, but they recombine in clouds, and we see these as fairly perfectly white in the right conditions. So I suggest our personal definitions of white are learned as an average of the signal balance we experience.

Edit: note that it should be possible to produce exactly the same "white" signal balance from your receptors with a spectrum of wavelengths very different from that from the sun.

 

[DaveC426913]

My point was that a colored shadow is not due to to actual wavelengths entering the eye.

As my example shows, sometimes it is exactly that.

But I grant that there are other scenarios where perception is the primary factor.

 

[brainpushups]

As my example shows, sometimes it is exactly that.

Fair enough. I suppose I could have chosen a better example like that of how the apparent color of a square on a mondrian appears different when the nearby colors are hidden from view. For example, a "green" square on a mondrian will appear green when viewed next to nearby squares in a very wide range of lighting conditions. Cover up the nearby squares and impose the same lighting conditions and one cannot discern that the color is green. Thus, wavelength of light (or combinations thereof) are not a sufficient condition for determining the color of an object. I would say that most scenarios are that in which perception is the primary factor. If wavelength of light was the primary determining factor of the color of an object then things would appear to change color as the lighting conditions changed and this is not how we experience color.

 

[DaveC426913]

Fair enough. I suppose I could have chosen a better example like that of how the apparent color of a square on a mondrian appears different when the nearby colors are hidden from view.

Yeah, I eventually realized what you meant.

 

[russ_watters]

What is exactly hitting the retina for us to perceive white light in a room radiated by light from a lightbulb?

Most of the other answers seem to be vaguely focused on this, but I don't see how this relates to the rest of the post, so I'll set it aside for now...

The light from the lightbulb caused by heat radiates wavelengths randomly through the space around it. But how could it be that the superposition of these random frequencies and wavelengths end up being close to white?

The simple answer is that the distribution of frequencies isn't random, but follows the black body curve:

And that when centered around different temperatures, you get slightly different distributions -- different shades of "white" from different lights:

 

[haruspex]

Most of the other answers seem to be vaguely focused on this, but I don't see how this relates to the rest of the post, so I'll set it aside for now...

The simple answer is that the distribution of frequencies isn't random, but follows the black body curve:
View attachment 239018And that when centered around different temperatures, you get slightly different distributions -- different shades of "white" from different lights:
View attachment 239019

And note that the whitest looking in that sequence (above the word "cool") roughly corresponds to the temperature of the sun's surface.