Modeling dynamic coupling in social interactions

Abstract

Whether negotiating the price of an item in a foreign marketplace or temporally coordinating actions within a musical ensemble, the basis of any social interaction must be the exchange of relevant information allowing for co-actors to become more or less aligned in body and mind. Although in humans, there are obvious daily examples of verbal exchanges, from infancy and beyond, we depend significantly on non-verbal exchanges. In most cases of social interaction, it is these non-verbal exchanges such as a hand gesture, a shift of gaze, a change in action timing, that allow us to predict, respond and adapt to one another, that is to become aligned. Both the experience and the process of becoming aligned is dynamic in nature, that is alignment happens and the degree of alignment changes across time. As such, an improved understanding of how we do things together, requires both a revised theoretical construct and dynamic timecourse models for predicting social behaviour between interconnected agents. The following is a summary of the work done as part of the interdisciplinary cooperation Group “Discrete and Continuous Models in the theory of Networks” at the Centre for Interdisciplinary Research (ZiF, Bielefeld University) to devise appropriate dynamic timecourse models. These models will be based on theoretical concepts of alignment as well as empirical testing of healthy individuals in which traditional measures of the degree of coupling will be used to compare instances where group size and richness of available social information is varied. Of specific interest will be phase transitions along a spectrum of self-organisation (negotiation, alignment, uncoupling). It is intended that the models will be used in conjunction with and compared with acquired and simulated data. Exploring the exchange of non-verbal information in a network of interconnected agents and bridging the gap between cognitive neuroscience and mathematics, this paper puts forward specific ways in which the use of dynamic timecourse modelling can further our understanding of the cognitive and neural underpinnings of social interactions. Specifically, findings will be translated to extend existing general sociological concepts of group behaviour and dynamics as well as specific applications in networks of individuals ranging from small groups like a string quartet or a football team to a large crowd at a political rally.