[ExI] are qualia communicable?

[Ben Zaiboc]

On 15/04/2023 23:01, Giovanni Santostasi wrote:
> Another even deeper mystery that the communicability of qualia is how 
> the brain creates an I. 

Oh, I thought that was simple. In it's essentials, anyway. I'm sure the 
details of the implementation are pretty complex, but the principle, as 
I understand it, is just that in amongst the many models we make, of the 
external world and other agents etc., there's a model of the agent doing 
the modelling. This model is referred to as 'I', just like the model of 
my cousin is referred to as 'Brian'. So when we say "Brian is going to 
the shops", we are making a prediction involving the 'Brian' model, and 
when we say "I am going to the shops" we are making a prediction 
involving the 'I' model (which of course encompasses the system doing 
the predicting). So you could call it a 'self-referential model'.

Or is this obvious and trivial, and you're talking about the details of 
how this is done?

If you mean the actual implementation, then I doubt anyone knows just 
yet. It's a general question about how the brain creates and manipulates 
models, especially models of agent systems. Probably quite high in the 
layers of abstraction, so analysing it in terms of neuronal connections 
will be difficult.

But once we know how the brain creates models in general, we'll know 
what an 'I' is, as it's just another model.

(Some models will be simpler than others, but going by how the brain 
works in general, and the massive duplication it uses, I doubt if a 
self-model will be that much different from a model of your room. Bigger 
and more complex, yes, but using the same principles).

Ben