The complexity perspective is probably most associated with the idea of complex adaptive systems (CAS) and the new modeling approaches that have developed from this way of thinking, agent-based modeling.

Complex adaptive systems (CAS) are characterized individual agents interacting with one another within a constrained environment such that properties emerge which inturn influnece the interactions that had originally given rise to them.  This influences the interactions, and these change the emergent properties, and so on and so forth. information is gathered and used by the system so that successful properties are selected and unsuccessful ones are extinquished, the complex system is also adaptive.

The individual agents, or particles in the case of physical systems, that interact with one another at what is called the fine-grained scale (Gell-Mann, 2002). These fine-grained interactions are characterized by a series of rules or constrains which determine the nature of how the interactions unfold. The nature of these rules, their specific instantiations, and how they are proscribed within the system are important elements in the analysis of these systems.

In early CAS modeling approaches (Holland, 1975) information gatheirng and use systems (Gell-Mann (2002) such as primative biological systems were simulated in an effort to explore how various sets of "rules of interaction" might cause the system to devolve into emergent properties. These most primative of agent systems were called cellular automata (Wolfram, 2000). Slightly more complex were systems where the "rules" were allowed to change through a process that simulated genetic mutation and recombination, simple systems of agent that included what were called genetic algorithms. Insights and issues raised by these models were the subject of John Holland's 1975 book Adaptation in Natural and Artificial Systems. Genetic algorithms were also the basis of important studies by Robert Axelrod (1985) with regards The Evolution of Cooperation

Human beings interact in much more complex ways, of course, and a new science is needed.  This is the area of Human Interaction Dynamics (HID) that is a focus of the Institute's research.