Emotionally Grounded Social Interaction

The question we address is the following: how do agents that are embedded in large societies — hundreds to millions of agents — focus the enormous potential for interaction with other agents towards the most productive encounters? The relevance of this question needs little argument. From a social studies viewpoint, it is tied to the understanding of a society's structure and how it — implicitly and explicitly, directly and through internalisation mechanisms — regulates daily interaction. From a multi-agent (MA) studies perspective, the issue is fundamental to the understanding of artificial societies. In particular, it enlarges the scope of behaviour control in agents, which is not only aimed at task achievement (e.g., finding food to survive), but at doing so in an evolving social context that, by the interdependencies it entails, makes every strict assumption of self-sufficiency a naive one. More technically, one cannot expect exhaustive exploration of potential encounters to be an efficient means of implementing large-scale agent societies. This is especially true for open MA systems — those in which the agents are being designed independently and without a predefined and globally known relationship towards the others. Finally, and here lies another important motivation for our work, such large-scale open agent societies will soon be of practical relevance — we predict that networks like WWW will evolve into large collections of software agents independently pursuing goals on behalf of their users, or trying to offer services to users and other agents. In pursuit of an answer this paper brings together two lines of research. One line has focused on understanding large heterogeneous societies in terms of separate but loosely coupled communities, and on developing technologies that could enhance social processes within such societies. This work on co-habited mixed reality (Van D. CAÑAMERO AND W. VAN DE VELDE 2 de Velde 1997) draws inspiration from anthropology, in particular concerning the role of bodily appearance and presentations to encode relationships of social stature, power and representation (see for instance Keane 1997). It studies the processes that result in the structural coupling between the different constituent communities of a complex society. The second line of research that feeds into this paper is on bodily-grounded emotions. This work stresses the role of affects as mechanisms for adaptation, both from an evolutionary perspective and in the sense that this term has in (Ashby 1952) — contributing to maintain the organism (a dynamic system) …