One important property of Complex Systems is the self-organization, which requires metacognition for self-regulation both from preservation and evolution perspectives.
Pavard and Dugdale advocate the use of structural and distributed approaches in the cognitive and social sciences, to subsidize the understanding of complexity in collective or organizational dimensions. They define four properties of a complex socio-technical system:
Non-determinism – it is impossible to precisely anticipate the behavior of complex systems, even if we know their functions and components, basically due to the cognitive dimensions involving their pluri-contextual variability, making it difficult to draw a standard stream of information in communication, as well as separate external and internal factors that influence it.
Limited functional decomposition - complex systems have a dynamic structure, it is difficult, if not impossible, to study their properties by decomposing them in functionally stable parts. This is because their permanent interaction with the environment and their self-organizing properties allow them to restructure, while people are not aware of the importance of pluri-contextual mechanisms on their performance.
The distributed nature of information and representation – they have properties comparable to distributed systems (in the sense of connections), that is, some of its functions may not be precisely located.
Emergence and self-organization – complex System have emergent properties that are not directly accessible (identifiable and anticipated) from the point of view of components, to the extent that these may not forestall knowledge regarding the operation of the components of these systems.