Rhythm as an ordered phase of sound: how musical meter emerges in a statistical mechanical model

We develop a model of musical rhythm and meter based on optimizing the trade-off between human psychological preferences for perceiving repeated patterns in time with a desire for variety and complexity. By mapping these competing preferences onto analogous quantities in statistical physics, we define an effective free energy which is minimized in the grand canonical ensemble. Using a mean field approximation, we observe phase transitions in the model from disordered events in time to orderings that closely reproduce those seen in music. We then compare the range of rhythmic characteristics predicted by the model to a dataset drawn from compositions by Johann Sebastian Bach, finding generally good quantitative agreement. The results provide a new lens through which to study musical rhythm, and a method for generatively producing rhythms.