Readers of my joculum blog will be familiar with my citations of Richard Lewontin's assaults on the sociobiologists and their inclination to create just-so stories to fill in for evidence when it comes to Darwinian explanations of the genetic bases of human behavior. (Lewontin, like Stephen Jay Gould, holds to a more nuanced explanation of the relationship between evolutionary structures and social structures.)
So I'm amused by my own just-so stories, as much as by Denis Dutton's just-so stories in The Art Instinct, which I reviewed semi-favorably, but with some very serious disagreements, on this blog some months back.
So bear in mind the ease with which "I can imagine that such-and-such happened" becomes "it might have happened" and then "it must have happened" and then "it most certainly happened."
But recalling from D'Arcy Thompson's On Growth and Form onward how the same basic natural forms are reshaped, recycled, and adapted to fit functions for which they originally never developed (I almost said "were intended"), I would not be at all surprised to find that some of our emotional reactions to color and shape are based on neuroreceptors that serve quite different purposes in other life forms. The resources of nature are immense, but they are not infinite. There is no intrinsic reason why we should not have complex emotional reactions to shapes and colors we were not meant to see (or more accurately, could not possibly see without technological assistance, such as coral reefs in the Pacific). The reactions could have (did have? no..."imaginably might have") been part of a complex biochemical response that evolved without reference to the use of these improbably pigmented mixtures of chemicals in far-distant and taxonomically independent species. When we encounter the stimuli, we have reactions that are not at all related to the functions of the shapes and colors in question, just as we have reactions of "cuteness" based on the visual cues provided by human infants needing nurturing, even when the creatures being viewed need no nurturing whatsoever.
There are directions in which all of this could be taken, but absent further experimental evidence I am going to say that I like my own fiction very much. And I think it has as much empirical evidence going for it as anything the sociobiologists ever fantasized as an explanation. Whether it has anything to do with why designers suddenly love improbable shades of blue and adopt them for things like airline tableware and seat cushions (or new-car interiors)...well, some just-so stories are intrinsically sillier than others.