At first I found this exercise very difficult. Sometimes I couldn't see any effect of the background on the colour of the smaller square. Sometimes all I could see was flashes of after-colour (usually of the complementary colour) surrounding the smaller square.
I wrote down what I thought, but I was very much not confident about it.
I moved on to the next part of the exercise.
I tried to guess what colours I was seeing, and in some cases this helped me to see very slight colour changes. I wrote them down but as I write this I am still unsure as to whether I am seeing what I am meant to be seeing. For some of them I didn't put down a colour but a comment about a change in brightness, if that seemed more apparent to me.
I tried to work out how these colour changes related to the relative positions of the two colours on the colour circle. I put my thoughts about this in visual form beside each one. After a while there seemed to be a pattern to the changes I saw, that the colour of the central square moved away from the colour of the background.
I then tried to find combinations of colours that might produce clearer perceived colour changes in the central square. So I used primary colours instead of the bright but non-primary colours I had tried before.
This produced much more obvious colour changes in the central squares. I tried the circle representations of the changes again and got a similar result to before.
Then I did the same exercise with mid-grey squares in the middle, which resulted in the greys appearing different in each one, their colour leaning towards the complementary colour of the background. It was so much clearer with the greys that I wondered why this hadn't been given as the first part of this section.
I hadn't thought before about the way we see colours being thoroughly dependent on the context, in an analagous way to the way we see shapes and lines being dependent on it. (I know this from trying to learn to draw.) That there is a big difference from what we think we see and what we actually see. And that when drawing something we have to switch our brains into a mode which allows it to represent what it actually sees as opposed to the illusion.
It seems that the same idea applies to colours - that it is important to reperesent the colour actually seen rather than the one which the labelling part of the brain tells me I see. I assume that this is going to take some work to find tricks to get myself into the right frame of mind, just as it did when I was learning to draw.
I'm looking forward to seeing how this works in practice in the next set of exercises, and slightly daunted as well considering how far I have to go to train my eyes