The goal of the Common Agent Theory (CAT) research program is to develop a broad, minimalist theory of agent and agent interaction behaviours. Development of this base theory will allow for a more unified investigation of common agent dynamics in a way that is generalizable across many different types of agents (e.g. human, animal, digital). At the same time, from a research perspective, the theory can act to highlight the underlying commonalities across existing agent behaviour theories and frameworks.
The motivation for the CAT research program, and the theory itself, comes partly from the conceptual work of the Human Exploitation Dynamics (HED) project. Work to describe simulation elements in a technology agnostic fashion led to an increasingly abstract view of agents and agent behaviours. This abstract agent perspective fits with a growing recognition of the relevance of agent behaviours “qua” agent behaviours, rather than agent behaviours serving only as representations of human or other organism behaviours.
The name of the theory, Common Agent Theory, and its goal of unifying existing agent theories, is directly inspired by the Common Model of Cognition (Laird, Lebiere, and Rosenbloom 2017), which seeks to integrate disparate theories of human cognition into one standard model or theory of cognition. This theory is a major focus of research at the Carleton Cognitive Modeling Lab, where I’m a visiting professor.