Simple models of network learning

In his talk, Simon would like to share two new models on which he is currently working. Both are small, simple and stylized representations. The first deals with questions of vicarious learning, transactive memory and trust in system of two agents (at a later date an N agent model will be tested). Trust is modeled as the belief that another person's advice can be relied on and this belief is seen as no different from other beliefs about the state of the real world. The second model builds on a prequel (Rodan ,2008), a model of network learning. Its focus is on 'defensive' learning heuristics in an environment in which some agents, not readily identifiable, employ 'tertius strategies' (Burt 1999, 2000) that makes it  problematic to rely on the information they  provide.