Consciousness is hot in science – and therefore cause for heated debates, conflicts and even conflicts of direction. In an open letter Last week, 124 neuroscientists and philosophers strongly opposed a theory that is steadily gaining attention, the ‘integrated information theory’ (IIT) of consciousness. This is “pseudoscience”, say the signatories, including big names such as the Americans Patricia Churchland and Daniel Dennett, who defend a purely physicalist view of consciousness.
The signatories believe the new theory, developed in 2004 by neuroscientist and psychiatrist Giulio Tononi, unscientific and unverifiable. They also fear risky social effects if it is widely embraced, for example with regard to abortion and euthanasia (they believe that it follows from IIT that plants, fetuses and coma patients also have consciousness).
Sympathizers of the theory see the action as grossly exaggerated and a misplaced accusation. The reason for the letter was reporting in Nature and Science about empirical support for IIT, but according to them those messages were not at all favorable.
This is also the opinion of David Chalmers, the Australian philosopher who put the ‘hard problem of consciousness’ on the map in the 1990s. He complains to X about the “disproportionate” letter. “Calling IIT pseudoscience is like dropping an atomic bomb on a local dispute,” said Chalmers.
‘Integrated Information Theory’ assumes that consciousness is a property of complex, integrated systems of information processing. In principle, this could also apply to physical systems other than the human or animal brain, for example even to plants. The theory has a mathematical elaboration, with phi as a symbol for the degree of information integration (and ultimately consciousness).
IIT is making waves and, according to physicist Sabine Hossenfelder, is currently “the most popular mathematical theory about consciousness,” she writes in Existential Physics (2022). But the theory has also been fiercely controversial from the start. Neurophilosophers criticize this as unverifiable or even as a form of ‘magical thinking’. According to some, IIT comes dangerously close to panpsychism (the idea that the entire universe is conscious), despite the limitation that it explicitly concerns complex information systems and not ‘everything that exists’.
IIT advocates dispute that the theory is untestable, but Hossenfelder points out that the calculations needed to phi of a system with today’s computers would take an impossibly long time, just to establish ‘consciousness’ in a worm. Moreover, according to her, simple information systems are conceivable that are sky-high phi being able to score without being aware of it.
IIT has broad appeal, but has to compete with other theories of consciousness such as HOT (Higher Order Theories), RPT (Recurrent Processes Theory), PPT (Predictive Processing Theory) and GNWT (Global Neuronal Workspace Theory).