It is a dynamic model, evolving to adapt to a problem. It is an hybrid model, between a semantic network and an artificial neural network. From the semantic network, it gets the symbolic nature of its nodes. From the neural net, it gets the activation propagation, through weighted links. This model is more or less like Everything engine.
One can say it works by nodes association (when a node is activated, and it is strongly bound to another one, it will activate this other node). Furthermore, building a Concept Network in a certain manner, one can say it works by symbols association an concepts emergence. The aim is obviously to activate the more relevant nodes in order to launch agents in charge of finding and creating instances of the Concept Network nodes into the Blackboard.
The initialization of the Concept Network is done by instances creation in the Blackboard of the nodes representing the problem to compute. Knowing that such instances creation activated their father nodes, the initialization problem is partially solved. The building of the Concept Network have to be done according to this.
It matches the Copycat's Slipnet.
BAsCET's model is a Concept Network.