In terms of evolution or brain science, why would people enjoy art, especially abstract art? Why would a human brain think an unnatural pattern of shapes and colors is something "good" to look at?

It can stir up interest. There is some art that the colors can make you happy, having it hanging in your apartment. Then some art stirs up drama. It gives you a feeling. A visual object that can stir up feelings.

I see nothing good in them. But it all depends on perception. Human beings are unpredictable and choices of all people can't be same. The rule is that, whatever you make, there will be at least a couple of  thousand persons in the world who will love it. So that's the secret. Do what you love.

This is a controversial topic with a rich history of discussion and some data. Here are a couple of ideas to inspire your own future research: Genetic Evolutionary Arguments:Art is a complex behavior just like any other and evolved for classic evolutionary reasons. In other words, genetics factors associated with producing art or enjoying art led to higher survivability. Note: this could mean that the art behavior itself somehow led to higher survivability OR that some other behavior caused higher survivability and also happened to lead to art. Art itself may not have increased survivability, but just "came along for the ride" because it correlated with other factors that did increase survivability (e.g., increased spatial reasoning). To read the most famous explanation of this idea that not all traits are associated with survivability, check out: The Spandrels of San Marco http://rspb.royalsocietypublishing.org/content/205/1161/581.full.pdf Culture Evolutionary Arguments: The existence of art is best understood in terms of cultural evolution, not the evolution of genes. Memes are behaviors or ideas that can be learned by imitation and therefore undergo their own evolutionary process independent of genes (for an awesome book on this idea check out: The Meme Machine by Susan Blackmore  http://100medo.com.br/documents/LIVROS/TheMemeMachine1999.pdf ). From this perspective, art was a random behavior that someone started one day by chance, and for reasons that have nothing to do with survivability, but a lot to do with the structure of human brains, more and more people started doing art, and the nature of what is "good" art evolved through its own evolutionary process independent of genetics. Neurobiology ArgumentsThese are not mutually exclusive from genetic or cultural evolutionary arguments. Rather, in either genetic or evolutionary arguments, the evolution of art is associated with specific and essential features of the human brain. A field called http://en.wikipedia.org/wiki/Neuroesthetics has been exploring how understanding the brain can help understand the nature of art. I think that these research efforts are probably most closely aligned with the OP's question, and I encourage you to explore further. ConclusionThere is no definitive answer at this time to how or why art evolved, or a clear understanding of why some art is considered "good", "beautiful", "interesting", by a consensus of people. In my opinion, cultural evolution theories and neurobiology are the most promising areas of insight for this question. thanks for the A2A

I found an interesting neuroscience article exploring the neural correlates of abstract art, in this case Kandinski paintings. I'm not a neuroscientist, but I found it pretty intriguing! Here is the link (may be behind a firewall):  http://ieeexplore.ieee.org/xpls/icp.jsp?arnumber=6033528 There is also a field called http://en.wikipedia.org/wiki/Neuroesthetics; the article has a lot of information there too.

The unusual patterns presented by abstract art elicit interest and curiosity, and those mental states are intrinsically pleasant for most people. Sometimes, in the artistic works that go beyond good to great, the perceptual effect stimulates an experience that takes one far outside the habitual familiar confines of cognitive experience, hard to describe, but it can be like taking a step outside oneself to peek into another world, a very pleasing experience.

The term "natural selection" is a misleading description of the process of evolution. Evolution is not a additive (selective) process but rather a process of deduction (de-selection); non-viable traits are eliminated by penalty of death. Traits that are carried forward aren't so much selected for as they are the traits that weren't selected against. This means evolutionary traits can be benign or even harmful to members of a species (but less harmful than an trait they are mutually exclusive with). Just because a trait is attributed to members of a species seeking out a certain stimuli or engaging in a certain behavior doesn't mean that trait is providing any biological advantage to that species, or that is related to another trait that is providing some sort of advantage. I think the speculation as to the evolutionary "advantage" of trait is fun to think about it, but we should always stop before indulging in this thought experiment with the intent to actually describe the natural world. There is a certain logic to describing sexual attraction as being a evolutionarily developed trait. But logic in-and-of itself doesn't establish a process of cause and effect. We have enough trouble defining traits, much less prescribing evolutionary causes and demonstrating them empirically.