Well, now we're all on the same page, and there's no disagreement. This could turn into a long conversation about explicit vs implicit communication. And we could trace all the way back and find out who misunderstood what, and how this argument came to be... :) Perceive and see are synonymous to me, "see" is "perceive visually"
Without doing a study, I'd say (good assumption) male population knows fewer adjectives for colours than female population (just the nature of guys liking ugly trucks, and girls liking dolls from the early age; girl toys being more colourful than guy toys). And I think there have been enough studies done to show that penises make creatures like different stuff than vaginas by default (mysterious, but appears to be true)
Of course I'm not talking about colour blind folks. They'll see according to their disability. Although it doesn't have to be a disability. I'm sure that if you raise a child under a red light, they'll be disabled and won't be able to see other colors even though they have the sensors to do it (brain has a tendency to connect nervously to stimulating sensors, so even though sensors are there, the brain will learn to ignore the input from the inactive sensors). Similarwise you can argue that Northern people are more sensitive to blue shades and southern people are more sensitive to red shades (probably true to a very small degree).
But as far as your study goes, it has very little to do with the rainbow (if anything at all). It's more of a memory/linguistic test. Like observing people draw, and people will draw the same thing differently. Some people can draw/paint photorealisticly, others symbolically, yet others suck completely :)
And if one puts a character and asks a group of people to draw that character. Some might see sad eyes, beautiful figure, porous skin, happy wrinkles, hump, limp, small nose, kissable lips.... And another guy barely took a glance at the figure and drew a figure stick. Is he wrong? Well depends on the context.... But overall, there maybe some true proportion in the picture, but human body is definitely not a line.
So if all those people want to mark black and white as a part of the spectrum, well it's just plain wrong. There's no black and white (or gray or brown), that means they haven't devoted enough memory space to accurately describe a rainbow. It's only as true as to say that a rainbow is many colors in a circular fashion around an axis of a light source (I don't think most people would think the geometry either). But what defines a rainbow is not that is is "many colors", but that it is a particular sequence of colors (all from 400nm to 800nm, whatever the range is), and measured by our instrument (eyes) with a high enough precision to tell at least a 1000 of them apart, and be able to name at least 20 into our crude languages.
It is also similar to the way a white man says that all chinese people look the same. These are all similar "lack of dedicated memory" scenarios. People can only memorize a particular amount of detail about a subject, and they start by memorizing the most general details to the most complex.
So what your survey seems to measure is the ignorance of the human brain (I'm using "ignorance" with no negative implications, being ignorant is an ability to ignore an uninteresting feature or a feature too complex, so it might need revisiting to memorize)...
I could mumble for pages about my fascinations with human brain.