Finite State Machines with Feedback: An Architecture Supporting Minimal Machine Consciousness

Abstract

Finite state machines with feedback present a novel machine model when considered under the scenario of cognitive computations. The model is designed in the spirit of automata theory and presents a mix of Alan Turing’s finite state machines and Norbert Wiener’s machines with feedback. For the model we define, what we call, minimal machine consciousness and machine qualia. The design of our model is lead by natural engineering requirements. Its properties are justified by the latest findings in neuroscience and by ideas from the classical literature of the philosophy of mind. For the model a test distinguishing minimally conscious machines from unconscious ones (“zombies”) on a given cognitive task is proposed. Our modeling supports the claim that consciousness is a computational phenomenon that is not just a matter of suitable software but also requires a dedicated architecture.