In cognitivism, manipulation of symbols became a central idea. In the 1980 an alternative framework emerged in network models. In network models, or connectionism, construction of knowledge does not involve manipulation of symbols. When the network is unstable (a need, a problem arises) information is gathered until the network is stable again (the solution is found). (Connectionism and the mind, an introduction to parallel processing in networks, William Bechtel & Adele Abrahamsen, 1998 Blackwell isbn 0-631-16577-0).
In my humble opinion this is an answer to the question of an agent in a network.
Graeme Ferris: ” … it does not address the dynamic nature of the entities, agents, humans, etc that are used to model the “nodes” in a network. …” . The dynamic nature is in the unstable state of the network. And of course humans in the network will feel the unstable state of the network.
The network is the agent. When the network is unstable, because a problem or a need is emerging in the network.
Than the network will be gathering information, sharing information, make connections to new nodes, until reaching a stable state in the network.
This unstable state of the network is tension, curiosity, feeling and emotion.
If I see a problem, my network will gather information, make me feel unstable, make me search for information.
After some time the information input and sharing and connecting is over. The answer is there. I feel stable.
Swedinbalchik: “… since in an early stage the coherence isn’t there (we have fragmented information elements). Learning is achieved when we pull together and connect the various information elements in such a way that it reflects our life experiences, circumstances or type of work currently involved in. …”
The we is not a little (wo)man in our heads, it is the network.
Compare Learner No agents but content creators?