Reversible Boolean networks. I: Distribution of cycle lengths (Q5932054)

This article discusses a class of time-reversible models describing the dynamics of \(N\) Boolean variables, where the time evolution of each variable depends on the values of \(K\) other variables. The authors introduce the concept of special points and the distinction between special cycles and regular cycles. The relation between special points and the properties of the cycles is demonstrated. They show that the numbers of special points and special cycles for each realization are proportional to \(\omega=2^N\), where \(N\) is the number of variables in the system. The correlation between typical cycle length and the \(K\) values of the networks is studied, and the authors find that the typical cycle length increases logarithmically with \(N\) when \(K<1.4\), exponentially when \(K>1.7\), and following a power law when \(K\) falls in between these two values.