What Is Color?

This year's "Flame Challenge" asks scientists to explain color in terms an 11-year-old can understand. The rules limit answers to either 300 words of text or a 6-minute video. 300 words is ridiculously short, so video is clearly the way to go. Of course, I'm not much of a video expert, but then, one of the finalists last year (when the question was "What Is Time?") was just a guy talking into a webcam, and hell, I can do better than that. So I did this:

(This is, obviously, why I was fooling around with looking at the spectrum of light from my laptop a little while back...)

The approximate text of the video's narration, if you don't want to watch it, is:

The question “What is color?” is actually pretty sneaky, because what we think of as color is a combination of two sciences: the physics of light, and the biology of sight. They interact in weird ways, making questions about color tricky to answer. There’s no better example of this than the color purple. Every kid in kindergarten knows that if you mix red and blue you get purple, but physics tells us that there’s no such thing.

In physics, what we describe as color has to do with the wave nature of light. If you shake a charged object or a magnet up and down, it creates a disturbance that travels through space and causes other objects to move up and down as well. We call this a wave in the electromagnetic field (which just means “a wiggling disturbance that affects electric charges and magnets”). That wave is what we know as light.

Physicists describe waves in terms of how fast they wiggle up and down—the frequency—or the distance it travels during one trip up and down, the wavelength. To a physicist, visible light is just a particular range of wavelengths—between about 400 and 700 nanometers, or a few one-thousandths of the width of a hair—and the color of the light is related to the wavelength. Different colors of light correspond to different wavelengths.

We measure light using a tool called a spectrometer that tells us how much light at different wavelengths something produces. A red laser produces a lot of light at 660 nm, and very little at other wavelengths. Green light has a kind of medium wavelength—around 540 nm—as we see with this green laser, and violet light has the shortest wavelength of anything we can see, around 400 nm. In physics, color means wavelength, and every color is its own unique wavelength. Violet light, like this 405-nm laser, is not a combination of red and blue, but its own thing.

So where do we get the idea that purple is red plus blue? Well, if we look at the spectrum of a picture of that violet laser, on my computer, we don’t see a sharp peak at a single wavelength, but a broad mush, with light at lots of different wavelengths.

So, why is that? That’s where biology comes in. We don’t have spectrometers inside our eyes to measure wavelengths; instead, we see colors of light thanks to special cells in the back of our eyes that send a signal to our brain when they detect light. There are three types of these, each sensitive to a particular range of wavelengths. There are “long wavelength” cells that are most sensitive to orange-red light, “medium wavelength” cells that are most sensitive to yellow-green light, and “short wavelength” cells that are most sensitive in the blue. When one of these cells gets hit by light in the right range, it sends a signal to the brain saying “Hey, I see some light!” The brain collects signals from all three types, and puts them together to decide what color light you’re seeing.

The ranges of these cells overlap, so yellow light will make both the long- and medium-wavelength cells respond, and green light will make both short- and medium-wavelength cells respond. If the long-wavelength cells respond more than the medium-wavelength ones, then it knows the light you’re seeing is more red than yellow. If the short-wavelength cells respond more than the medium-wavelength ones, then it knows the light you’re seeing is more blue than green.

This is why the color-adding trick works. You can fool the eye into thinking it’s seeing a particular wavelength of light by combining other wavelengths in just the right way. Green light makes both the short-wavelength and medium-wavelength cells in your eye respond, but so does a combination of blue light (which makes the short-wavelength cells respond) and yellow light (which makes the medium-wavelength cells respond), which is why we say that blue and yellow add to make green. With the right mix of colors, you can fool the brain into thinking that it’s seeing any wavelength of light that you like.
The displays used in TV’s, computer monitors, and smart phones use three sources of light: one mostly red, one mostly green, and one mostly blue. When my computer tries to reproduce the color of the 405nm laser, it uses a mix of these; mostly blue light, with some red and even green mixed it.

This color mixing also allows us to “see” colors that don’t exist in physics, like purple. There isn’t a single wavelength of light that corresponds to what we see as purple—rather, it’s a mix of mostly-red and mostly-blue light. We can even see this by mixing together the red and blue lasers we looked at before: in the region where the two beams overlap, our eyes see an entirely new color, one that physics tells us doesn’t really exist.

So, what is color? It’s what happens when the physics of light combines with the biology of sight to make the world that we see, in all its glory. Even if some bits of that exist only in our minds.

There are pictures galore in the video, but I'm not uploading them all here. Who do you think I am, Ethan Siegel? The "featured image" up top, showing overlapping red and violet lasers, is as much as you're going to get...

More like this

Green is a lie.

"What colors does a Dog see?" should at least make a cameo appearance!

Physics tells us Purple doesn't really exist? A basis of the infinite-dimensional space of EM field configurations that contains various kinds of Purple is just different from the all-conquering and ever-useful frequency basis.

By Peter Morgan (not verified) on 03 Mar 2014 #permalink

Swing and a miss. It might be just me, but I would avoid using units, numerical measurements, graphs and words like "spectrometer" if your target audience are 11 year-olds. Flirting with the wave nature of light is a challenging task to convey to an 11 year-old in under 300 words.

By William Hendrixson (not verified) on 03 Mar 2014 #permalink

I'm find with "spectrometer", since you define what it is. "Disturbance" might be a bit too much, and I'd want to know what a nanometer was at that age.

I guess the other question this would leave me with is why those proteins in our eye are responsive to certain frequencies.

Color is a subjective sensation that cannot really be explained. Sure you can talk about light and frequency and so on but that is only a small part of the story. So light of a certain frequency hits your retina, and is converted into electrical impulses, but how are those impulses mapped to colors? What is it that makes us experience 450nm light as blue? Do we even experience blue in the same way? Obviously colorblind people don't. Can you imagine a new color you have never seen before? How would it feel to have one's colorblindness cured at age 20?

The interface between external objective reality and it's subjective reflection is so poorly defined and understood. Where does one end and the other begin? Where do our sensations come from? Is the pool of possible sensations limited and independent of biology or is it unbound, our particular senses being tailored by our cellular structure to the specific needs of evolution? Will it be possible one day to implant completely novel senses into our consciousness?

The puzzle of color is inextricably linked to the puzzle of consciousness.

Funny, I was thinking about colour last night after listening to a BBC radio programme about dragonflies. The insects apparently have a much more sophisticated colour vision system than we do. My daughter asked me if this meant that they could see more colours than humans could. I suppose the answer is yes, in the sense that I'm guessing that they are more sensitive to tiny changes in the spectral composition of colour. Two blues that look identical to us, would look different to a dragonfly.

I suppose the answer is yes, in the sense that I’m guessing that they are more sensitive to tiny changes in the spectral composition of colour. Two blues that look identical to us, would look different to a dragonfly.

A couple of weeks ago, in one of the Uncertain Dots hangouts, I mentioned that one of the post-docs working with sodium, back in the day, spent a good chunk of an afternoon bringing people into look at colors. And while there's a broad swath of wavelengths that people will shrug and call "green" before you get to "blue-green" or "yellow-green," the range for unqualified "yellow" is really narrow. Which probably had to do with falling right between the peaks of two of the color receptors, allowing maximum sensitivity to small change.

Channelling my inner 11-year-old, I get lost pretty early on:

If you shake a ... magnet up and down, it creates a disturbance [we call] a wave... That wave is what we know as light.

I just shook a magnet up and down. No light!

More (I hope) helpfully, from trying to explain things to my own kids I'd say it's critical to use analogy, because that means using words they know and ideas they can relate to. But then you need to stick within the analogy (as far as possible) and to using the words they know, not the new words they haven't had a chance to memorize (and taking into account that even an expression like 'the cell responds' might not be intuitive to a child). Where there is a new word that's unavoidable, it needs to be signalled and re-taught each of the first few times its used ('so then the atom - remember, that's the thing we said was a bit like our lego bricks - sticks to another atom - another "lego brick"...').

You're teaching an unfamiliar concept that builds on unfamiliar mechanisms, and using a vocabulary that's completely alien to do it.

Imagine someone trying to explain (to you) a grammar function in a foreign language that is fundamentally different from your own. Start by assuming that you have no frame of reference - say, that you're used to a language with almost no cases, like English, and this is a language with a sophisticated case system, like Czech. Now imagine them introducing you (once!) to the grammatical terms used *in that language*, and then going on to explain the complicated point while using only those terms.You'd get lost. Kids will here, too - it is almost exactly as foreign!

By Pan Outeast (not verified) on 04 Mar 2014 #permalink

So, on one of my kids' song tapes from long ago, an annoying guy warbled "Red Orange Yellow Green Blue Purple" and I always had to loudly interpose INDIGO VIOLET over the last word. Now I feel justified.

Our senses are our body and brains interpretation of signals from the outside world. For sight, these are electromagnetic radiation that emits or is reflected from objects. Electromagnetic radiation consists of things such as radio waves, micro waves, x-rays, and visible light. The thing that differs between the different kinds is their wavelength- or the distance between the peak of two adjacent waves. Visible light is just the section which we can detect with our eyes- roughly 400 and 700 nanometers (10^-9 meters).
However, our eye does not directly detect the wavelength of light that enters our eyes. Our eyes have 3 types of sensors can detect 3 ranges of visible light- S: 400–500nm, M: 450–630nm, L: 500–700nm. They respond more to light in the middle of the range than the ends. Generally S matches as blue, M as green and L as red. When we see light that only activates S, then the brain knows it is from the early 400s, and will interpret it as purple or blue depending on how many S sensors activate. If M strongly responds, L moderately responds, and S doesn’t, then the light is around 530 or so and is cyan or green. Those two examples are however of individual light waves. When there is a range of light waves, or a combination of overlapping light waves, then we get the common knowledge of red + blue = purple or blue + yellow = green. Our brain is just interpreting an object that emits both light that matches red and blue as purple, a mix of blue and yellow light as green.
Color, is then, our brains interpretation of which 3 of these sensors are responding and how much they are responding.

There, <300 words and gives the main points with fairly common words. Still somewhat technical though. Submit it if you want.

Rereading that, it could use some indentation and pictures (was mainly trying to squeeze it to the 300 word limit). For a video, I'd definitely add some pictures diagrams for A) what a "wave" looks like and what's the wavelength B) a rainbow that shows where the 400 to 700nm are C) some visual examples of identifying colors and mixing colors. D) Also for an actual lesson: what "white and black" are and possibly some things regarding depth perception.

Tested this on my 12 year old. He thought it was interesting. He was particularly pleased with the idea that purple doesn't really exist.

He found the description of light a bit simplistic, to give you a sense of his sophistication -- he fed you the words "wavelength" and "frequency" for example.

Hi Chad.

This video ia amamzing!