with Anil Seth

connectionism?

Seth spent time on the study of artificial life.

definition of consciousness: the existence of a subjective experience. more folk definition.

definitions evolve with our scientific understanding; we don't need a rigorous definition to move forward, as long as we're not talking past each other.

phenomenal consciousness vs access consciousness.

there are some things we have a folk understanding of but are totally intractable to study scientifically. e.g. memory: we have an understanding of it as a single thing, but in neuroscience it appears to be many different things.

consciousness may be more than one thing. we may be misled into viewing it as more singular than it truly is.

the formulation of the hard problem rests on a conceivability argument for zombies, i.e. entities that are functionally the same as us but are not conscious.

Harris critiques the analogy of consciousness to the problem of describing life: unlike consciousness, everything about life can be described in extrinsic, mechanistic terms. Seth: perhaps this is only apparent in hindsight. throughout the history of science, phenomena have seemed inexplicable before turning out to be explicable. it's an open question whether the analogy to the problem of life is false. we shouldn't decide a priori that there's something really different about consciousness.

Seth: we may be asking too much of a theory of consciousness.

our current guess about what's conscious: definitely mammals, maybe birds. this is all just extrapolation for now.

octopus is a great example of evolutionary convergence. they may be conscious too, in a different way from us.

precautionary principle: given that conscious suffering is really bad, we should assume more entities are conscious than less, to be careful.

potential dimensions of consciousness: the level of consciousness, the contents of consciousness, and the conscious self.

consciousness vs wakefulness.

anasthesia awareness: when you are aware while under anesthesia.

a crucial experiential difference between anasthesia and sleep is that with sleep, when you wake up you feel that some time has passed. with anasthesia, you disappear and then instantly you are back. complete oblivion. at the same time, it's impossible to distinguish between oblivion and a failure of memory.

there's never such a thing as "raw" perception. perception is always a balance between external stimulation and internal prediction or inference about what is there. this is the "controlled hallucination" theory of perception.

Bayesian brain or unconscious inference from Helmholtz. perception works more from the inside out than from outside in. example: there's a constraint built deep into our visual system that light always comes from above; this forces us to interpret shadows and curves in a particular way.