Subject: the color constancy problem

I've read a little about "color constancy" in humans, having to do with our 
ability to see the same color of an object under different light conditions 
(either more or less light or even different wavelengths of light, I think).  
Apparently this "isn't supposed to occur" and we should see an object's color 
change much more dramatically, for example, when a cloud passes overhead 
(according to http://www.cs.sfu.ca/people/Faculty/Funt/funt).  My question is 
this: How important is this function to humans, and what kind of "dramatic 
color changes" would we see if we didn't have color constancy?  

Re: the color constancy problem