A theoretical computer science perspective can deepen understanding of consciousness. The Conscious Turing Machine (CTM) is a simple, substrate-independent computational model for defining and exploring consciousness, analogous to how the Turing machine models computation. The CTM is not a model of the brain or cognition but offers a framework using computational complexity theory and machine learning. This paper introduces the approach, demonstrates its possibilities, and aims to stimulate research on consciousness from a theoretical computer science viewpoint.
A network of brain regions called the default mode network breaks down during anesthesia and after brain damage causing disorders of consciousness. The neurochemical reasons for this breakdown were unclear. Using functional MRI, researchers found that the ventral tegmental area, a dopamine-producing brainstem region, disconnects from key default mode network nodes (precuneus and posterior cingulate) during both propofol sedation and disorders of consciousness. Stronger connectivity between the ventral tegmental area and these nodes was associated with a more awake-like configuration of the default mode network. In patients with disorders of consciousness who later improved behaviorally, this connectivity increased toward healthy levels. In a separate group of traumatic brain injury patients, the drug methylphenidate significantly strengthened this connection. The findings suggest that dopamine modulation may be central to maintaining consciousness.
The conscious experience of seeing motion in a particular direction is linked to the activity of specialized neuron clusters in the brain's motion-processing area. Using high-resolution 7-tesla fMRI, researchers identified clusters of neurons in the human motion complex that responded preferentially to either horizontal or vertical motion. When participants viewed an ambiguous display that could be seen as moving either horizontally or vertically, their reported perceptions alternated every 7 to 13 seconds. The activity in the horizontal- and vertical-tuned clusters dissociatively reflected which direction the person perceived at any moment, even though the visual input remained constant. These clusters were organized in columns, with stable direction preferences through the depth of the cortex.