Abstract In this paper, we introduce a novel method for classifying and computing the frequencies of movement modes of intra and interspecific dyads, focusing in particular on distance-mediated approach, retreat, following and side by side movement modes. Besides distance, the method includes factors such as sex, age, time of day, or season that cause frequencies of movement modes to deviate from random. We demonstrate and validate our method using both simulated and empirical data. Our simulated data were obtained from a relative-motion, biased random-walk (RM-BRW) model with attraction and repulsion circumferences. Our empirical data were GPS relocation time series collected from African elephants in Etosha National Park, Namibia. The simulated data were primarily used to validate our method while the empirical data analysis were used to illustrate the types of behavioral assessment that our methodology reveals. Our methodology facilitates automated, observer-bias-free analysis of the locomotive interactions of dyads using GPS relocation data, which is becoming increasingly ubiquitous as telemetry and related technologies improve. Our method should open up a whole new vista of behavioral-interaction type analyses to movement and behavioral ecologists.