An important target of the present research is to model “comfort” in the human brain, whereas traditional models of brain and neural networks concentrate on memory and learning.<?xml:namespace prefix = "o" ns = "urn:schemas-microsoft-com:office:office" />

We propose that the composite cycle model of two standard neural networks derived from statistical physics reveals the mechanism of comfort-generating. The composite cycle model demonstrates mysterious temporal oscillations, with frequencies related to the super-magic numbers, including the golden ratios, which give living beings comfort. Because the numbers can also be observed in music, sympathetic resonance occurs between inside and outside of the brain, which gives comfort. Then we conduct more detailed examination of the relation between the model’s outputs and six kinds of comfortable chords of music. By optimizing some parameters in the model, we confirm conformity between all the six kinds of the ratios of the chords and the outputs. This result will lead to an artificial genius to produce creative works including comfortable music on the computer.