Music, new aesthetic and complexity

Abstract

This paper illustrates an algorithm to generate a complex acoustic stimulus whose statistical properties are as close as possible to the non-stationary dynamics revealed by the current analysis of the electro-encephalogram activity of the human brain. Thus, the composition is driven by crucial events, namely renewal non-Poisson events with an inter-time distribution density ψ(τ), which is an inverse power law with index μ, fitting the condition 1 ≤ μ ≤ 2. We find that the music composition is more attractive when we fill the time region between two consecutive crucial events so as to enhance the leading role of μ. In all cases the spectra markedly depart from the ideal 1/f condition, thereby suggesting a shift from the 1/f noise perspective of the pioneer work of Voss and Clark to the Zipf's law perspective advocated by more recent work on music composition. © 2009 ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.