← Back to context

Comment by lolinder

2 years ago

To be relevant to blue/green wouldn't they need to not just collect a boolean but collect the type of colorblindness?

It gets complicated if the goal is perfect accuracy. Cone sensitivity also varies on an individual basis even for color-normal people. Worse, the transfer function of the eyeball also varies with age as your lens yellows and internal fluid clouds a bit. Even holding those constant, brains do a lot of processing that maps what your eyes can physically capture into perceived colors, which are significantly influenced by upbringing.

Plus, screens and ambient lighting. It's a lot of variables.