I believe in the idea of the rainbow.
And I’ve spent my entire life trying to get over it.
– Judy Garland
Blind Spots Great and Small was about the fact that we all have a pretty large blind spot in each eye. These blind spots are actually bigger than the fovea, the tiny area in each eye that picks up half the visual input to the brain. So blindness is wider than sharp focus!
But when we look round the room we are never aware of the blind spots because some part (or parts!) of our brain fill in the gap. It’s a bit like a Photoshop job: the perceptual system completes the picture with something that matches the surroundings well enough to stop us noticing.
It’s well known that this amazing filling-in system can match colours and textures. But after writing the previous post about this I started to wonder: what if the colour was non-uniform? For example the background might change smoothly, say from light to dark. Will my brain fill in the missing bit of the gradient where the blind spot is, or will I see a step change?
First here’s a warm-up example, just in case you haven’t tried these bind spot experiments before or are out of practice.
Close the right eye and fix the left eye on the X. Move your head nearer and further from the screen and when you get the distance right the O will be in your blind spot and disappear.
You can do it with the other eye of course: close the left one instead and fix on the O. When you get the correct distance the X will vanish.
Notice that in both cases you are not actually aware of the blind spot: the O or the X is no longer visible but where it was (is?) you just see a continuous background.
Now let’s try that again but make the background a smoothly varying grey scale. Close the left eye, fixate on the X, and move you head closer and further from the image as before, until the O disappears. What do you see?
The perceptual system seems able to fill in the gradient very convincingly, at least mine does!
If the gradient is coloured it seems to work just as well.
What if we have a gradient between two colours? I’ve picked two that I hope will still work if you are colour blind.
In that case, I reasoned, perhaps a multicoloured gradient would be filled in as well. And what is the best-known multicoloured gradient? A spectrum of course.
So here goes! If you are colour-blind I apologise in advance because the experiment may not work for you.
Close the right eye, fixate on the X. When the O is in the blind spot what do you see?
When I first tried this I was confidently expecting my brain would fell in the missing part of the spectrum. But alas no!
I see a spectrum but with yellow and even orange missing. Green sort of blends into red, although the transition is a bit indistinct and I can’t tell exactly it looks like.
Here’s a flipped version in case you want to try with the other eye.
What’s happening? I’ve been thinking about this for days but I still don’t have a really satisfactory explanation of why the yellow vanishes. The best theory I have is this.
The simple gradients are processed by the earlier stages of the visual system and it’s there that the filling in takes place. However these lower–level stages don’t know about ROYGBIV. They can’t tell that Orange and Yellow “ought” to come between Red and Green. We are aware of it cognitively but the higher levels where that knowledge lives is a different part of the brain from where the filling of the blind spot is handled.