Comment by embedding-shape
5 days ago
I'm fairly sure we can measure human "sensation" as in detect physiological activity in the body in someone who is under anesthesia yet the body reacts in different ways to touch or pain.
The "feelings" part is probably harder though.
We can measure the physiological activity, but not whether it gives rise to the same sensations that we experience ourselves. We can reasonably project and guess that they are the same, but we can not know.
In practical terms it does not matter - it is reasonable for us to act as if others do experience the same we do. But if we are to talk about the nature of conscience and sentience it does matter that the only basis we have for knowing about other sentient beings is their self-reported experience.
We know that others do not experience the exact same sensations, because there are reported differences, some of which has been discussed on HN, such as aphantasia. The opposite would be visual thinkers. Then you have super tasters and smellers, people who have very refined palats, perhaps because their gustary and/or oilfactory senses are more heightened. Then you have savants like the musical genius who would hear three separate strands of music in his head at the same time.
Absolutely - I have aphantasia myself, and did assume for 40+ years that my experience was like everyone elses, but I didn't want to make the argument more complex. It's indeed correct that we have often assumed we think the same way but have reasonable reason now to think that isn't actually true. But it still feel reasonable to accept that we're probably close enough. But still, we absolutely can't prove it.
You can measure model activity even better.
How do you know that model processing text or image input doesn't go through feeling of confusion or excitement or corrupted image doesn't "smell" right for it?
Just the fact that you can pause and restart it doesn't mean it doesn't emerge.