The latest attempt to use science to explain art pulled people from different cultures, and used fMRI to evaluate their brain patterns as they made aesthetic judgments about paintings and pieces of music (the original study can be found
here). Aiming to "formulate a brain-based theory of beauty," the study's authors from University College London walk into an arena fraught with tension - after all, countless hours have been spent composing tenuous (and racist, sexist, ethnocentric) scientific criteria for evaluating beauty.
Twenty-one subjects ranked pieces on a scale of beauty, from 1-9, and the study's authors used these rankings to sort pieces into the categories of beautiful, ugly, and indifference-inspiring. Attempting to side-step entrenched art-historical debates about cultural conceptions of beauty (typically opposed to traditional, humanist claims of universally appreciable beauty: see
Winckelmann), the study's authors left the actual evaluation of beauty up to each individual subject. Leaving aside the question of what makes the beautiful beautiful, the study's authors seem to ask whether the human brain even recognizes beauty as a phenomenon.
Apparently, they found that the human brain responds to "beautiful" stimuli very differently from "ugly" stimuli. While I can't speak to the science of their findings, it's important to note (as Katherine Harmon
does) that the study has significant limitations. It doesn't attempt to define what made subjects rank a painting more beautiful; nor does it bear on the role of beauty in evaluating art by other criteria. Notably, subjects made their judgments on 16-second flashes of paintings and pieces of music. "Slow art," or forms of beauty that take time to emerge, disappear from the study's purview.