http://papers.nips.cc/paper/6222-brains-on-beats.pdf
In the field of machine learning such as image recognition, image recognition is performed by imitating the brain. With the technology called CNN, there are layers that are learning information that is easy to see, such as the color of the image and what is drawn, and layers that are learning information that is difficult to understand at first glance (painting style, etc.) .. That is, it is clear that they are recognized separately.
In this paper, we investigated how the brain processes music.
As a result, it became clear that high-level processing (style, etc.) and low-level processing (melody, etc.) are clearly separated in the brain.
In other words, it was found that AI was created by imitating humans, but the human brain actually recognizes things in the same way. (There are multiple ways of recognition)