| Like the Roman god Janus, the IDA project has two faces, its science face and its engineering face. Its science side fleshes out the global workspace theory of consciousness and cognition, while its engineering side explores architectural designs for software agents that promise more flexible, more human-like intelligence within their domains. We can expect the architectures and mechanisms that underlie consciousness and intelligence in humans to yield software agents that learn continually, that adapt readily to dynamic environments, and that behave flexibly and intelligently when faced with novel and unexpected situations.
How Minds Work is intended to give a detailed picture of several mechanisms of mind combined to produce a conceptual model of cognition in the context of a human-like software agent, IDA. But why should you want to engage yourself with this tutorial? The burning question that drives my research, and that of many others, is how do minds work? Understanding the conceptual model that underlies the IDA model buys me something. Given a question about some cognitive process, I can most often ask how does it work in the model, and get an answer. The hypothesis thus generated is that it works the same way in humans. This hypothesis may or may not be correct. That’s left for the cognitive scientists and the neuroscientists to sort out. But I have an answer that gives me at least the illusion that I understand how minds work, or at a minimum, how they could work. It’s heady stuff, and to me, worth the effort of understanding the IDA model deeply enough to get answers.
This tutorial is intended to bring about just such an understanding in the reader. The diligent reader should emerge from this tour with a clear understanding of the workings of the IDA model. That is, he or she should carry away a clear picture of how this model says that human minds work, not only in their function, but also something of their mechanisms.