A new method for evaluating the degree of association between the behavioural patterns of 2 individuals is described and illustrated. It is proposed discrimination should be made between the unit for sampling (sample interval) and the unit for analysis. The length of the sample interval affects analysis accuracy, while the length of the analysis unit affects the analysis scale. A new "moving analysis unit (MAU) method" is proposed to divide an observation period into analysis units. The MAU method, with various lengths of analysis units, allows for various aspects of interaction between 2 individuals. As a similarity coefficient between the 2 temporal patterns, Iwao's ω (Iwao 1977) is proposed. Investigation of nonparametric confidence intervals of ω is performed with the bootstrap (Efron 1979), which is also applied for obtaining the null hypothesis distribution of ω. The MAU method, ω and the bootstrap are applied to the analysis of both insect and human communication.
ASJC Scopus subject areas