3 June 2014

Post date: Jun 5, 2014 12:13:53 PM

Babak talked about variations of agents presented in the literature ranging from simple ‘entities’ with passive features to very sophisticated actors with cognitive functioning. Using a compilation of various characteristics of agents from the literature (see, for example, Wooldridge & Jennings, 1995; Franklin & Graesser, 1997) different groupings of some of these characteristics were used/ presented to particularly differentiate between ‘proto-agents’ and ‘agents’ with cognitive capabilities.

Then, using one of the definitions of ‘weak’ (upward causation) and ‘strong’ (downward causation) emergence (proposed by Muller, 2002, discoursed by Dessalles and Phan, 2005) it was discussed how cognitive characteristics of agents are essential in order to have ‘strong’ emergence dynamics in the model (but not so for the ‘weak’ emergence). Similarly, it was argued that the real ‘self-organised’ dynamism in a model can be realized by agents with learning and adaptability capabilities (idem for ‘complex adaptive systems’).

It was also briefly discussed which ‘basic’ agent’s characteristics (along with other factors) are essential to make ABM a suitable modelling approach (see, for example, O’Sullivan et al., 2012).