Is consciousness computational?
computational in the sense of discrete turing machines
Case yes: - slicing problem with water computers - counting computations problems: - assume i run a consciousness computation on a large computer. now i copy this process and execute it again, increasing the total cpu needed for this. i doubled consciousness. - now i introduce caching of certain computations shared between the two programs. how many consciousnesses are there? Case no: - this implies one of two things: - a simulation in which the observable properties of consciousness (such as discussing consciousness) evolve are not computable - this implies that physics is not computable. This could for example be true if physics is intrinsically continuous and holds infinite information - but is is a bit weird that we can simulate physics very well in many cases, and we can also get a lot of AI with discrete computations, but our conscious experience depends on those errors that our simulators do - the observable properties of consciousness don't imply the existence of consciousness, and there is no actual relationship between discussing consciousness and being conscious
Is physics computational?
Define a physics computation problem as an initital state and a time T, and the solution to the problem is the probability distribution over possible states of the system at time T.