Introduction to Special Issue on Collective Effects of Human Behavior

Abstract

The cognitive capacities of humans are highly developed. Consequently, the complex patterns of social interactions observed among humans are usually attributed to their sophisticated abilities, such as planning, intention, and universal knowledge. The study of human behavior, therefore, rarely uses the approach developed in artificial life: that interactions among simple units may unexpectedly result in complex structures by selforganization. This differs from studies of social organisms of lower cognition, such as social insects, to which this approach has been applied often [2]. However, also in humans complex patterns of interaction may emerge through self-organization. This has been demonstrated, for instance, in one of the earliest works in this field by the Harvard economist Thomas Schelling [7]. He was interested in ghetto formation in large cities. In a model, Schelling shows that, even if the desire to be close to individuals of one’s own type is slight, its behavioral consequences may increase at a group level, leading to unexpected “macropatterns.” Schelling’s example has been followed by a number of studies along these lines, in economics [3], in politics [1], in anthropology [5], and in linguistics [4]. Therefore, this approach appears to be useful for the study of collective behavior in humans. This kind of modeling is called synthetic. It integrates rules at the level of individuals with behavior at a group level, which is typical of studies in artificial life [6]. In this issue of Artificial Life, it is applied to a broad range of aspects of collective behavior of humans. In the first article, Janssen and Jager define artificial consumers who use different decision processes depending on their degree of uncertainty and satisfaction with products. They may, for instance, repeat their previous decision, imitate the choice of others, or decide themselves. The authors show that, apart from the process of decision, the social network of the agents also influences product diversity and market stability. In the second article, market stability is studied if firms communicate over different distances. Distance of communication has recently been getting larger, which is partly due to our improved transportation systems and communicatory tools (telephone, fax, e-mail, e-business). Louzoun et al. model this by using an extension of the LotkaVolterra model of predator-prey systems in which firms operate over different distances. In this model, exchange at a global level leads to the emergence of a single dominant market system. This system is prone to collapse, because it is not buffered by other market systems. It appears that communication between firms over a medium distance (smaller than global) maximizes both wealth and stability. Most of our communication is expressed via language. A frequently posed question is