Dynamics of citation distribution Original Research Article

We study the citation dynamics of scientific publications over the years. We propose a simple cellular automaton model featuring a combination of two distinct mechanisms, i.e. the random assignment and the preferential attachment, to investigate the dynamics of journal citation. Different from most previous studies focusing on highly cited papers, we analyze the time evolution of the entire citation distribution. Empirical data can be well reproduced by numerical simulations. Within the linear regime of the Cited Half-Life, a steady accumulation of citations can be expected. Moreover, within this linear regime, the ratio between the above two mechanisms is a constant. Besides the average citation represented by the Impact Factor, such a constant ratio can also be a characteristic of the journal.