Covering topics such as the mind–body problem, the observer problem, neural coding, emotions, symbol grounding, associative processing, inner speech, and situational awareness, the book explains how mental phenomena can be analyzed, modelled, and implemented in machines. It then develops these concepts into concrete engineering approaches through the HCA associative architecture and its hardware proofs, demonstrating precursors of conscious perception in robotic systems.
This monograph will interest researchers, graduate students, engineers and cognitive scientists working on advanced robotics or AI, as well as philosophers of the mind delving into AI and machine consciousness.