Building life-like 'conscious' software agents
Stan Franklin · 2000
Here we will briefly describe action selection and language generation mechanisms in two ‘life-like ’ software agents, CMattie and IDA, and discuss issues that bear on the topics of architectures for behavior control, interdependencies be-tween emotions and goal-based behaviors, and coordination of scripted and improvised behaviors. These agents are life-like in the sense of interacting with humans via email in nat-ural language. They are ‘conscious ’ only in the sense of im-plementing a psychological theory of consciousness [3,4]. At this writing we are exploring the transition from scripted lan-guage production to more improvised speech generation. We are also investigating deliberative behavior selection mecha-nisms whereby alternative scenarios are produced and evalu-ated, and one of them chosen and acted upon. 1. Agents: autonomous, cognitive and ‘conscious’ The rise of the web has spawned a use of the word ‘agent ’ in other than its more common meaning as in ‘travel agent ’ or ‘insurance agent ’ or ‘real-estate agent’. In this new context an agent is a piece of soft-ware, a computer program, that in some sense acts on its own, typically in the service of its user, a per-son. Here I’ve chosen to define the technical term ‘au-tonomous agent’, at least partly to avoid the seemingly endless debates over exactly what is an agent. An au-tonomous agent, in this paper, is a system situated in, and part of, an environment, which senses that envi-ronment, and acts on it, over time, in pursuit of its own agenda. It also acts in such a way as to possibly in-fluence what it senses at a later time [14]. These au-