In their Orch OR theory Penrose and Hameroff propose that consciousness results from sequenced interaction with space time curvature through networks of microtubule structures in the brain. These structures enable the entanglement of quantum states contained in multiple spatially separate neurons and provide a means to amplify the quantum world to the macro world. This mechanism opens the door for the non-computational properties of orchestrated quantum wave collapse to influence the information processing activities of the brain. The implications of this gives rise to many questions as to what role the quantum world might play (if any) in macro-level processes of decision making, creativity, and other aspects of cognition. The proponents of the theory have suggested that the information contained in the quantization of space time curvature is an essential element of cognitive function. Opponents have argued that the brain has evolved specifically to eliminate quantum level “noise” from its information processing mechanisms. In order to further the understanding of these questions, I propose an artificial system that has the capability to be “plugged” and “unplugged” from the influence of the underlying quantum universe. This capability provides a controlled environment for studying and thinking about the relationship between quantum phenomena and high-level cognitive behavior. Additionally, because it does not rely on distributed computation for decision making and knowledge representation the number of entangled quantum states required for implementation is significantly reduced. This puts the possibility of realizing a working system in reach with current technology. The system I propose is based on a traditional AI blackboard architecture. It utilizes a control algorithm where the evolution of a quantum system is guided by the symbolic computation of the knowledge sources and the blackboard. The overall system operates in repeated sequences of updates to the blackboard, a mapping of the blackboards symbolic state to a set of entangled quantum states, and objective reduction of the quantum system to determine the control output. One way to think of this is that the macro-level symbolic computation orchestrates the computation of a synthetic quantum system. To uncouple the system from the quantum world the quantum control algorithm is simply replaced by the traditional blackboard control algorithm. Although I focus on blackboard architectures, I also show how several other popular AI / Cognitive architectures could be augmented in similar ways. Unlike the distributed processing and storage characteristics of the microtubule networks of the brain, this system localizes the interaction between the macro and quantum worlds to a very specific interface and makes explicit the division of labor between the two.
