Before asking this very dismissive question, did you pause to think about how you might design an experiment to determine a rough count of how many colors people can discriminate between? Because it took me all of 30 seconds to do so, and I'm just a graph theorist. I don't know shit about anatomy or color spaces.

I'd use a 2-panel diagram. On each round, the panels are assigned random colors. The subject is asked "are the panels same, or different?" Some proportion of the time, they're identical. If they just mash "different" every time, you'll see that in the data and you'd reject those subjects -- or accumulate those false guesses into an uncertainty measure. If they hit 'same' on different colors... make a graph with edges between those colors. Some time after the graph is connected, try and find a minimum vertex cover. The order if that vertex cover should be a reasonable estimate of the number of colors a person can see.