A commentary to a paper of Baars, Journal of Consciousness Studies, 4 (4) 1997, 322-234.

GLOBAL WORKSPACE AGENTS

Stan Franklin,

Institute for Intelligent Systems,

Abstract: In the target article, Baars has offered both a theory of consciousness and a strategy for scientifically testing the theory. This commentary is intended as an addendum. I'd like to suggest implementing global workspace agents as both an additional strategy toward scientific testing, and as a means of fleshing out the theory.

Though comprehensive, Baars' theory is, as a theory should be, quite abstract, offering general principles and broad architectural sketches. Questions of architectural detail, that is of just how functional components fit together and who talks to whom, are left open, as are questions of mechanisms, that is of how these components do what they are claimed to do. For example, in Baar's presentation the various types of contexts (perceptual, conceptual, goal contexts) are lumped architecturally. Though distinguished functionally, their architectural relationships, as well as their mechanisms, are left unspecified. As a good theory should, this one raises as many questions as it answers.

Let's explore autonomous software agents as sources of possible answers to these questions. An autonomous agent (Franklin and Graesser 1997) is a system situated in, and part of, an environment, which senses that environment, and acts on it, over time, in pursuit of its own agenda. It acts in such a way as to possibly influence what it senses at a later time. In other words, it is structurally coupled to its environment (Maturana 1975, Maturana and Varela 1980). Biological examples of autonomous agents include humans and most animals. Non-biological examples include some mobile robots, and various computational agents, including artificial life agents, software agents and computer viruses. We'll be concerned with autonomous software agents, designed for specific tasks, and 'living' in real world computing systems such as operating systems, databases, or networks.

Such autonomous software agents, when equipped with cognitive (interpreted broadly) features chosen from among multiple senses, perception, short and long term memory, attention, planning, reasoning, problem solving, learning, emotions, moods, attitudes, multiple drives, etc., are called cognitive agents (Franklin in press). Cognitive agents can play a synergistic role in the study of human cognition, including consciousness. Here's how it can work.

Global workspace theory is more than a theory of consciousness, it's a theory of cognition, even a theory of mind. Minds, in my view, are control structures for autonomous agents (Franklin 1995.) A theory of mind constrains the design of a cognitive agent that implements that theory. While a theory is typically abstract and only broadly sketches an architecture, an implemented design must provide a fully articulated architecture, and the mechanisms upon which it rests. This architecture and these mechanisms serve to flesh out the theory, making it more concrete. Also every design decision taken during an implementation constitutes a hypothesis about how human minds work. The hypothesis says that humans do it the way the agent was designed to do it. These hypotheses will suggest experiments with humans by means of which they can be tested. Conversely, the results of such experiments will suggest corresponding modifications of the architecture and mechanisms of the cognitive agent implementing the theory. The concepts and methodologies of cognitive science and of computer science will work synergisticly to enhance our understanding of mechanisms of mind. I have written elsewhere in much more depth about this research strategy (Franklin in press), which I've called Cognitive Agent Architecture and Theory (CAAT).

An attempt at implementing a global workspace agent in pursuit of the CAAT strategy is underway. Its first phase is to build Virtual Mattie, an autonomous software agent that 'lives' in a unix system, communicates with seminar organizers and attendees via email in natural language, and composes and sends seminar announcements, again via email, all without human direction (Franklin et al, 1996). VMattie, now designed and in the coding stage, implements about forty percent of Baars' global workspace theory. The second phase will add the missing pieces of the global workspace theory, producing 'Conscious' Mattie.

In VMattie, Baars' "vast collection of unconscious processes" are implemented as codelets in the manner of the Copycat architecture (Hofstadter and Mitchell 1994, Mitchell 1993). Working memory consists of two distinct workspaces (yielding an hypothesis). Perceptual contexts include certain nodes from a slipnet type associative memory à la Copycat, and certain templates in workspaces. The node type perceptual contexts become active via spreading activation reaching a threshold (another hypothesis). Several nodes can be active at once, producing composite perceptual contexts (another hypothesis). Baars says that "[o]ne of the remarkable features of conscious experiences is how they can trigger unconscious contexts that help to interpret later conscious events." The VMattie architecture fleshes out this assertion with mechanisms. Goal contexts are implemented via an expanded version of Maes' behavior nets (1990). Again they become active by having preconditions met and exceeding a time variable threshold (another hypothesis).

In 'Conscious' Mattie, the spotlight of attention will be added. Coalitions of codelets will compete for attention via Sloman's measures of motive (1987): importance, intensity, urgency, and insistence, distress producing, pleasure producing (another hypothesis). Associative learning, implemented via Jackson's pandemonium theory (1987), gives a mechanism for learning requiring "conscious access to what is to be learned." Learning becomes less of a "magical process." Baars seems rightly amazed at the capacity and the recognition ability of human memory. The implementation of this faculty in 'Conscious' Mattie's architecture is via Kanerva's sparse distributed memory, again giving an explicit mechanism, hopefully subject to experimental testing.

Though more could be said, I hope I've made clear the kind of synergistic role the design and implementation of cognitive agents can play in the study of theories of mind, their architectures and mechanisms.

References

Franklin, Stan (1995). Artificial Minds. Cambridge, MA: MIT Press.

Franklin, Stan (in press). Autonomous Agents as Embodied AI, Cybernetics and Systems' Special issue on Epistemological Aspects of Embodied AI.

Franklin, Stan and Graesser, Art (1997) "Is it an Agent, or just a Program?: A Taxonomy for Autonomous Agents," Intelligent Agents III, Berlin: Springer Verlag, 21-35,

Franklin, Stan, Graesser, Art, Olde, B., Song, H., and Negatu, A. (1996) "Virtual Mattie-an Intelligent Clerical Agent." AAAI Fall Symposium on Embodied AI.

Hofstadter, D. R. and Mitchell, M. (1994), "The Copycat Project: A model of mental fluidity and analogy-making." In Holyoak, K.J. & Barnden, J.A. (Eds.) Advances in connectionist and neural computation theory, Vol. 2: Analogical connections. Norwood, N.J.: Ablex.

Jackson, John V. (1987), "Idea for a Mind," SIGGART Newsletter, no. 181, July, 23-26.

Maes, Pattie (1990), 'How to do the right thing', Connection Science, 1:3.

Maturana, H. R. (1975). "The Organization of the Living: A Theory of the Living Organization." International Journal of Man-Machine Studies, 7:313-32.

Maturana, H. R. and Varela, F. (1980). Autopoiesis and Cognition: The Realization of the Living. Dordrecht, Netherlands: Reidel.

Mitchell, Melanie (1993), Analogy-Making as Perception, Cambridge MA: The MIT Press.

Sloman, Aaron (1987), "Motives, Mechanisms, and Emotions" From Cognition and Emotion, Lawrence

Erlbaum Associates Limited, (3) 217-233.