Chromatic aberration in the human visual system and how our brain deals with it

Hey everyone,

I am currently taking a class at Hopkins called the Visual System, and one of the first things we discussed this semester was the anatomy and phototransduction circuitry of the human visual system. One of the things we talked about was chromatic aberration, which is the inability of a lens system to bring 3 wavelengths of light (R/G/B) into focus at the same point.

Many people like to talk about lens aberrations when comparing different lenses, but have you ever thought about the chromatic aberration present in the human retina? I thought that some of you might be also interested in learning a bit about how our visual system deals with chromatic aberration.

Plus, I had a test earlier this week, so I typed out most of this stuff as review anyway. So I figured, why not post it on the forums and see if anyone finds it interesting


First, some basics. People with normal color vision have three populations of cone photoreceptors, each with a different photopigment within them. There are long-wavelength (called L or red) cones, medium wavelength (Called M or green) cones, and short wavelength (called S or blue) cones. The blue cones have a max absorption at around 440nm, while the green and red cones absorb best at 533 and 565nm, respectively.

Herein lies an inherent problem. With a simple optical design (such as our eye), it is not possible to focus 3 wavelengths of light to a single point. This is called chromatic aberration, and it has the effect of making some wavelengths appear blurry since they are not all focused perfectly onto the retina (or sensor, if we're talking about cameras). Lens manufacturers use sophisticated optical designs and special lens elements (such as fluorite) to deal with these aberrations, but it is simply not practical to build a lens system with 10 or more special lens elements in the human eye (imagine how much your eyes would bulge out of your head, haha). So, how does the human eye, with its single lens, deal with chromatic aberration?

1) The human retina uses long wavelength light as a guide. These long wavelengths are focused perfectly on the retina. As a result, the slightly shorter wavelengths corresponding to green will be focused very slightly in front of the retina. This is not a big issue because the peak absorbances of the L and M cones are relatively close together. However, the peak absorbance of the S cones are far apart from the other two, so they really take a hit here. Short wavelengths (i.e. blue) and are focused considerably in front of the retina as a result. So, instead of blue light being focused onto the retina as a sharp point, they appear more like blur circles. It's not hard to see how that is detrimental to visual acuity. So what happens to the short wavelengths?

2) The area of your retina specialized for high acuity vision (the macula) is covered with a yellowish pigment. This can be thought of as a UV filter of sorts, since it essentially prevents most short wavelength light from reaching the areas responsible for precise vision. In addition, the center of the fovea (fovea is center of the macula and the region of highest visual acuity) is stuffed full of L and M cones and has no blue cones. The retina is essentially working around the chromatic aberration issue by ensuring that short wavelength light is not used for precise vision.

3) So what is the purpose of the blue cones? Even though they are not used for precise vision, they contribute to our ability to see color. In fact, the signals from the S cones are kept separate from the other two all the way to the primary visual cortex in the back of your brain.

So there you go. The structure of our retina and the processing circuitry present in our brain elegantly take care of the issue of chromatic aberration. And our visual system doesn't even need fluorite glass to do it

Hope you guys found this interesting in some sense.

Interesting thanks for sharing

For some reason when wearing eyeglasses and in darker environments, I can clearly see color fringing in areas of high contrast (e.g. the boundary between a window and the wall). It's unfortunate for eyeglass wearers.

I wonder if this is a function of astigmatism?

Probably too many beers

In all seriousness, I wear glasses, have fairly high astigmatism but never noticed the phenomena you just described. I'll have to pay attention.

You'll particularly notice it towards the edges of the lens (haha even glasses have poor corner sharpness).

If you look real close in your left ear, there is an adjustment setting for each set of cones. I just spent an hour with an opthamologist, and he's got everything balanced now. Since f/stop is continuously variable, he gave me some medication that will allow ISO adjustment. Standard eye ISO is 25, so if I need to capture images at ISO 400, I take 2 pills one hour prior. Shutter speed varies by time of day. Late at night, the shutter seems to close without warning, however I get to view several movies while in this state.

This is a great thread. Nice to know all that.

Hmmmmm, that could help explain why I have problems getting a sharp images at gigs where they use all blue lights for part of the show. My macula just doesn't have the right sensors.

hollis_f wrote in post #11101075
Hmmmmm, that could help explain why I have problems getting a sharp images at gigs where they use all blue lights for part of the show. My macula just doesn't have the right sensors.

You got it. Visual acuity under blue illumination is reduced, especially if you are trying to look at something more than a few feet away. This reduction stems from peripheral location of the short wavelength cones and the chromatic aberration that affects the short wavelength light reaching the retina.

A number of years ago, we had an ice storm that left over an inch of ice on everything. I looked out to my back yard and up at the power pole in the next yard. For the first time, I actually saw "St. Elmo's Fire", as the transformer and the wires out about 6 feet either side were glowing purple. I raced back in the house and grabbed my trusty A-1 and Vivitar 120-600 zoom. I mounted everything on a tripod and took a full roll (36) worth of shots. All were time exposure on bulb.

When I took the film to the pro lab the next morning, I told them what I had shot. Despite their best efforts, all I got was a bunch of dark rectangles. The violet shade of the electicity was out of the range of recognition of the Kodak 100 film I was using.

Oh well, film has limits in viewing toward the high UV part of the spectrum just like our eyes!

you have to update the humans firmware version :>

Interesting information. Thanks. This means also rather than making expensive lenses that corrects CA, you can let the sensors do all the work? Obviously the sensor design has to change, but thta might be the future.

Just to show you how well the human body is made....there was something about the white balance to. the perception of the white modified by the brain. Like a white car on a sunset appears orange but the brain knows it's white...a camera won't make the difference and the car would be orange on the photo.. something like that

jfbrodeur wrote in post #11104036
Just to show you how well the human body is made....there was something about the white balance to. the perception of the white modified by the brain. Like a white car on a sunset appears orange but the brain knows it's white...a camera won't make the difference and the car would be orange on the photo.. something like that

If I understand it right, doesn't it only appear "not-white" because we are later viewing the photo under different light?

Very interesting, thanks for sharing.

996gt2 wrote in post #11101674
You got it. Visual acuity under blue illumination is reduced, especially if you are trying to look at something more than a few feet away. This reduction stems from peripheral location of the short wavelength cones and the chromatic aberration that affects the short wavelength light reaching the retina.

In presence of blue light only, I'd expect the eye to focus properly (i.e. no L-guide, no M-guide, fallback to S)... am I expecting too much ?