Authors:
(1) Lenore Blum (lblum@cs.cmu.edu);
(2) Manuel Blum (mblum@cs.cmu.edu).
Table of Links
2 Brief Overview of CtmR, a Robot with a CTM Brain
2.2 Conscious Attention in CtmR
2.3 Conscious Awareness and the Feeling of Consciousness in CtmR
2.4 CtmR as a Framework for Artificial General Intelligence (AGI)
3 Alignment of CtmR with Other Theories of Consciousness
4 Addressing Kevin Mitchell’s questions from the perspective of CtmR
7 Appendix
7.1 A Brief History of the Theoretical Computer Science Approach to Computation
7.2 The Probabilistic Competition for Conscious Attention and the Influence of Disposition on it
5 Summary and Conclusions
In this chapter we have presented a brief overview of a simple formal machine model of consciousness, known here as CtmR. The TCS perspective has influenced the design and definitions of CtmR, and conclusions we have drawn from the model.
Although CtmR is inspired by the simplicity of Turing’s formal model of computation and Baars’ global workspace (GW) architecture, our formalization is neither a Turing Machine nor a standard GW model. Its consciousness (access and phenomenological) depends on having more than a global workspace.
Importantly, CtmR also:
1. interacts with its outer world via input sensors and output actuators;
2. has the ability to construct models of its inner and outer worlds;
3. has a rich internal multimodal language, Brainish; and
4. constantly updates its states via predictive dynamics (cycles of prediction, testing, feedback and learning), all while operating under resource limitations (time and space).
CtmR is not a model of the human or animal brain, nor is it intended to be. It is a simple machine model of consciousness. Nevertheless, at a high level, CtmR can exhibit phenomena associated with consciousness (blindsight, inattentional blindness, change blindness, body identity disorder, phantom limb syndrome, …), and aligns with and integrates those key features from main theories of consciousness that are considered essential for human and animal consciousness.
The CTM model demonstrates the compatibility and/or complementarity of those theories. It supports (the credibility of) our claim that a conscious AI is inevitable, because it is clearly buildable and arguably a basis for consciousness.
Finally, the development of CtmR is a work in progress. While we have worked out many details of the model, there is much left to develop. More specifics will appear in our upcoming monograph.
Our goal is to explore the model as it stands, determine the good and the bad of it, and make no unnecessary changes to it.
6 Acknowledgements
We are grateful to Michael Xuan for his immense encouragement, and to UniDT for their long-term support. We appreciate numerous discussions with and helpful critiques from friends and colleagues: Johannes Kleiner, Ron Rivest, Hy Hartman, Alvaro Velasquez.
This paper is available on arxiv under CC BY 4.0 DEED license.