Conscious Turing Machine
Explanation
The Conscious Turing Machine (CTM) is a formal architectural proposal for consciousness articulated by Lenore and Manuel Blum (Turing Award 1995), published mainly in a paper in PNAS in 2022 and developed in multiple subsequent talks and articles. The Blums, very respected theoretical computer scientists, applied their expertise in computational theory to produce an abstract but mathematically rigorous model of a conscious system.
The CTM takes explicit inspiration from the Global Workspace Theory (Baars, Dehaene) and adds computational-theory precision. Its structure includes: multiple specialised processors (like the unconscious processes in Baars), a central scenario (analogous to the global workspace), competition between processors to access the scenario (mediated by chunks with attention weight), broadcast of content from the scenario to all processors, updating of processors based on what is received.
The model mathematically formalises these processes: each processor produces chunks with weight; a selector (attention mechanism) chooses the chunk with the highest weight and publishes it on the stage (the global workspace); the stage is read by all processors in the next cycle; the system iterates. The Blums argue that this simple design explains many properties of consciousness: unity of the conscious field at each moment, temporal seriality, flexibility, capacity for reflection on one's own contents.
A distinctive feature of the CTM is its emphasis on what Lenore Blum calls the feeling of self: the system can report on its own states (meta-cognition), experiences its contents as its own, can have thoughts about thoughts (self-awareness). The Blums present this not as phenomenal consciousness necessarily, but as a set of functional properties typically associated with consciousness in humans.
The CTM dialogues with theories of consciousness (GWT, IIT, higher-order theories) but maintains the rigour of a formal specification. It has generated debates: Is the CTM conscious? Are its chunks qualia? Could it be implemented and verified empirically? The Blums present their model as a blueprint that, if well implemented, would give rise to functional consciousness; whether also phenomenal is an open question.
For the theory of consciousness, the CTM represents the most serious effort by theoretical computer scientists to give rigorous structure to the problem of consciousness. It is particularly valuable for its mathematical clarity, for its explicit dialogue with neuroscientific theories (GWT), for opening verifiable implementation possibilities. Critics note that formalisation does not per se solve the hard problem: that a system has the components of the CTM does not guarantee that there is something it is like in it. But as an attempt to make the theory of consciousness more formal and verifiable, the Blums' proposal is a milestone. It clearly articulates what consciousness as computational architecture might mean and offers a testbed for future debates on conscious AI, one of the most urgent questions of the twenty-first century.
Strengths
- Simple and elegant mathematical formalism.
- Reproduces key phenomena of consciousness (attention, uniqueness, sleep).
- Clear continuity with GWT and the Turing tradition.
- Explicit proposal of computable consciousness.
- Productive framework for debates on artificial consciousness.
Main critiques
- Does not solve the hard problem: identification with qualia is a postulate.
- Strong functionalism debatable.
- Mapping with neural data still preliminary.
- Ambitious philosophical position not universally shared.