I think it’s useful mostly because of the extra pegging it allows for. Remembering something abstract like a number or tone contour is rather hard, but if you have something concrete, like a colour, this makes it easier to create mnemonics, which can be great for remember the pronunciation of characters you don’t read aloud very often. I think colours are still too abstract, but the step to something like elements (blue = water, red = fire, etc.) is pretty small. I wrote more about this here if you’re interested!

…and the MT tone colours are different again to those listed above! (1st: green; 2nd: blue; 3rd: red; 4th black; none for 5th…)

From my experience of a tone-deaf person trying to learn the tones, colors proved to be very useful when drilling the correct tone pitch in texts. Its a one less middle-step for my brain go through, hence faster and more fluent my production.

Personally I think it is not too good to use colors for remembering Chinese characters tones. Now let me tell you why I think so. Chinese is a language generated from 412 syllables adding 4(5) tones makes it around 1600. The problem I encounter when trying to recollect a specific character tones is that there are too many homophones, extreme amount compared to my native language. The other issue I see is that Chinese written discourse and Chinese spoken discourse are very dissimilar as to structures and characters used.

Chinese learners need the tones mostly for production, speaking. Reading, even when subvocalized, needs pinyin first, tones are secondary. For understanding and reading (subvocalization included) texts tones are marginal.

Does using colors to represent tones help learning the tones? I would argue it does not. In the sense that recollection of specific tone-character combination is not increased. However it helps me to find and practice tone pitch, focus more on tones and diction than the message.

This wasn’t meant to be an historical overview, but what is your “it works well” based on? Do you mean that it works well for you, or that maybe he has some reason to think it works? I doubt there was any research into this that long ago, but I could of course be wrong. As the paper I cited indicates, though, students thinking that colours work doesn’t make them work! The group that used colours thought well of them in the post-experiment survey, yet they performed worse than the group that didn’t use colours. I’m not saying colours never work for anyone, but that the evidence so far seems to point that they don’t help on average.

I have added Hanping! Sorry for always forgetting to include it, so thanks for reminding me!