Integrated information theory
Explanation
Giulio Tononi, an Italian-American neuroscientist, has developed since 2004 one of the most mathematically precise theories of consciousness: Integrated Information Theory (IIT). Its central thesis: consciousness is integrated information. Not just any information, but the kind that a system sustains as a unified and differentiated whole, irreducible to its parts.
The amount of consciousness a system has is measured by a quantity called Φ (phi), which quantifies how much integrated information the system generates beyond the sum of its parts. Systems with high Φ (such as certain cortical circuits in the awake state) are highly conscious; systems with low Φ (such as a massive hard disk, which has a lot of information but little integration) have low or no consciousness.
IIT starts not from the brain but from phenomenology. Tononi identifies "axioms" of conscious experience: it exists (it is real for the subject), it is compositional (it has distinct parts and aspects), it is informative (each experience is one among many possible ones), it is integrated (it is unified, not decomposable) and it is exclusive (it is definite, with a specific scope). Each axiom translates into a physical postulate that a system must satisfy to be conscious.
A notable consequence of IIT is that consciousness is ubiquitous and graded: any system with Φ > 0 has some degree of consciousness. This potentially includes thermostats, plants, or (more controversially) aggregates such as countries or planets. IIT is thus a mathematically precise form of panpsychism (or more accurately, panintegrationism).
The theory makes empirical predictions. The cerebellum, despite having many neurons, has low integration (relatively independent modules); therefore, according to IIT, its damage does not abolish consciousness. The cortex, with its high integration, is consciousness-supporting. Anaesthesia and deep sleep dramatically lower Φ. The Perturbational Complexity Index, derived from IIT, is used clinically to measure levels of consciousness.
IIT is probably the most controversial and discussed theory in the field. Its virtues (mathematical precision, quantitative predictions, derivation from phenomenological axioms) are notable. Its problems (computational intractability of Φ for large systems, panpsychic implications that some consider absurd, debates about its falsifiability) are equally serious. But few theories have so marked the contemporary debate.
Strengths
- Phenomenologically grounded axiomatic framework.
- Potentially quantitative measurements of consciousness.
- Applicable to non-biological systems (animals, AI).
- Explicitly addresses the qualitative structure of experience.
Main critiques
- Calculating Φ is unfeasible for real complex systems.
- Panpsychic implications (any system with Φ>0 would have some consciousness).
- Some axiomatic aspects are contested.
- Difficult decisive falsifiability.