Bringing Networks to the Next Level: Self-modeling Networks for Adaptivity and Control of Mental Models

Networks provide an intuitive, declarative way of modeling with a wide scope of applicability. In many cases also adaptivity of a network plays a role, which easily leads to less declarative and transparent forms of modeling by using algorithmic or procedural descriptions for the adaptation processes. This chapter addresses this by exploiting the notion of self-modeling network that has been developed recently. Using that, adaptivity is obtained by adding a self-model to a given base network, with states that represent part of the base network’s structure. This adds a next level to the base network, resulting in a two-level network. This construction can easily be iterated to obtain more levels so that multiple orders of adaptation can be covered as well. This brings networks to a next level in more than one way. In particular, a three-level self-modeling network can be used to integrate dynamics, adaptivity and control in one network. It is shown how this can be used to design network models for mental model handling.