Human Perception of the Soundscape in a Metropolis through the Phenomenology of Neural Networks

The aim of this work is providing a quantitative description of the evolution from a hi-fi to a lo-fi soundscape in a metropolis, according to the definition given by R. Murray Schafer. Our theoretical approach has a statistical mechanical foundation and is based on order/disorder transitions. We will show that a system formed by C citizens and by the sound sources related to the K neighborhoods of the metropolis is thermodynamically equivalent to the fully connected network of interacting citizens described by the Hamiltonian of the Hopfield model. The application of statistical-mechanics models to the soundscape will confer an original interpretation to the concepts of ferromagnetic and spin-glass phases: such phases will be associated to different models of society (ordered vs. frustrated system). Transitions will be regulated by the variation of the control parameter, here defined as the ratio between the number of sound sources and the number of citizens.