The topic is interesting. In the current issue of 'Lettre International' Germany there is a very long article about the latest findings on how music and language are processed by the brain. For copyright reasons I can quote the article only very briefly in excerpts in the ASR.
The title: Wilde Welt der Musik
... a neuroengineer at Columbia University, and others have subsequently demonstrated that certain neurons in the cortex respond to certain consonants and syllables in our utterances. If one observes which neurons are activivated, it is possible to reconstruct the sentence that a subject has just spoken. It even predict which note of a song a person is preparing to listen to: The brain can recognize which melody is being played, and therefore seems to activate the relevant neurons in advance.
...
... the path that music takes through the brain is never straight. It's less like the sound that runs through a speaker wire and more like the data that flows through the Internet: Every phrase, rhythm, and sound is dissected, distributed, and reassembled in an infinitely complex network. It's even difficult to isolate the signal.
... As the synthesizer standing a few rooms away was developed, Sulzer told his students that the sounds he produced were mathematically too perfect to be musical. A pure sine wave is boring as hell" he said. Circuits had to be built in to contaminate it a bit. As it turns out, a little noise is essential to the sound of any instrument. Woodwind instruments croak, bows grind, voices hum and strings blur in overtones and harmonics.