5 ms·

What if consciousness is a spectrum? What if consciousness is the logical end-game starting with the ability to selectively attend to and respond to stimuli?

What if consciousness is a spectrum? What if consciousness is the logical end-game starting with the ability to selectively attend to and respond to stimuli? That's attention schema theory. And in such a theory, dogs are pretty far along the spectrum, because they (probably) have the ability to model others' internal states. After that, it's just a question of the complexity of the modeling. https://www.nytimes.com/2014/10/12/opinion/sunday/are-we-really-conscious.html https://www.nytimes.com/2014/10/12/opinion/sunday/are-we-rea... https://www.theatlantic.com/science/archive/2016/06/how-consciousness-evolved/485558/ https://www.theatlantic.com/science/archive/2016/06/how-cons...

tsimionescu 6y ago
Sure, that's a possibility. It is basically a truism that we can quantify the "consciousness" of a system to some extent, where we can say rocks have none, single cells (or even cell mechanisms) have the most basic kind, and more complex organisms have increasing amounts of this. Humans obviously have more of this semi-quantifiable "consciousness" described above than plants or fish, and it is likely that we have the most out of all living organisms we know; but identifying this quantifiable property with our subjective experience of consciousness is nothing more than conjecture, though. The OP's conjecture that consciousness is just our ability to model others' behavior reflected back towards ourselves is similarly just a conjecture. We are still entirely missing any tools, even at a theoretical level, that would help bring our subjective experience of consciousness into the realm of the externally observable, so we can only speculate for now. Perhaps in time AI research will change course again and try to return to such questions, instead of trying to become better at differential diagnosis, rote translation etc. (extremely useful goals, mind you, but scientifically uninteresting).

tsimionescu 6y ago

Sure, that's a possibility. It is basically a truism that we can quantify the "consciousness" of a system to some extent, where we can say rocks have none, single cells (or even cell mechanisms) have the most basic kind, and more complex organisms have increasing amounts of this. Humans obviously have more of this semi-quantifiable "consciousness" described above than plants or fish, and it is likely that we have the most out of all living organisms we know; but identifying this quantifiable property with our subjective experience of consciousness is nothing more than conjecture, though. The OP's conjecture that consciousness is just our ability to model others' behavior reflected back towards ourselves is similarly just a conjecture. We are still entirely missing any tools, even at a theoretical level, that would help bring our subjective experience of consciousness into the realm of the externally observable, so we can only speculate for now. Perhaps in time AI research will change course again and try to return to such questions, instead of trying to become better at differential diagnosis, rote translation etc. (extremely useful goals, mind you, but scientifically uninteresting).