How do Individual Team Members Regulate their Behavior to Achieve Spatiotemporal Collective Behavior?

Introduction This study aimed to analyze processes of the emergence of collective behavior patterns. Collective behavior, considered as self-organized, emerges from individual activities that interplay as the activity unfolds. The purpose of the present study was to explore how co-agents actively (co-)regulate their involvement to contribute to the emergence of collective behavior on-site. We did not consider, as often assumed in complex-systems theories, that co-agents couple locally per se, but that the way in which they couple has to be described empirically (Bourbousson & Fortes-Bourbousson, 2016), especially by exploring how they can act by grasping the overall collective behavior they contribute to. To this end, we aimed at describing the variety of ‘modes of regulation’ achieved by team members during a soccer game. To conduct such an ambition, we adopt an epistemological approach respectful for how human actively regulate their actor/environment coupling (AEC), that was the radically enactive perspective (De Jaegher & Di Paolo, 2007). In this perspective, sense-making is assumed to be central in delineating the dynamics of the AEC, and phenomenological experience of the actor is seriously considered in the study designs. Methods One team of expert soccer players (i.e., national level) participated in this study. We video recorded a complete official game using a drone. A sequence of ten minutes was selected, based on which phenomenological data was collected for all of players through individual self-confrontation interviews. The verbal data obtained were transcribed verbatim, from which we reconstructed the visual perception and the associated behavioral adjustment as players can comment it for each instant of the activity under study. Then, a thematic analysis (Braun & Clarke, 2006) helped to identify patterns of meanings within verbalization data, leading to the characterization of various modes of regulation. These modes were then re-positioned within the course of the experience of each teammate, so that co-regulation modes could be identified at each instant of the game. Results The results showed the amount of typical patterns of regulating activity carried out by participants, going from a mode of ‘local couplings’ to one of ‘couple through grasping a global picture’. In the collective behavior, the results showed that all the team members were able to share the same mode of regulation at the same time. Discussion This results showed that one or several members switched to another mode of regulation, thus helping to avoid the disruption of collective behavior. In other words, the robustness of collective behavior is associated to an individual flexibility. Together, the results provide news opportunities for coaching teams, as it suggests avenues for rethinking how swarming intelligence is conceived in multi-agents systems.