this post was submitted on 26 Oct 2022
5 points (85.7% liked)
Science
13883 readers
24 users here now
Subscribe to see new publications and popular science coverage of current research on your homepage
founded 5 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
In my opinion one of the key prerequisites for solving the hard problem is coming up with a theory that explains the necessity of consciousness.
Personally, I've always been partial to the idea that what we experience is effectively a simulation synthesized by the brain. This makes a lot of sense from thermodynamics perspective. Working on an internal model is much cheaper than parsing sensory input directly. Having an internal model that gets synced up with sensory input at intervals is efficient, and it explains a lot of things like sensory illusions, dreams, effects of psychedelic drugs, and so on.
The theory that consciousness is needed as a mechanism for synthesizing memories in order to to project into the future provides a potential insight as to why such a simulation is useful for an organism to develop in order to increase its chances of survival.
I imagine that the work on AIs is the most likely path towards figuring out the hard problem. We can test such theories by constructing neural networks that produce the desired behaviors and then seeing whether the observed behaviors match the predictions.
It's of course impossible to say definitively whether a system has qualia of experience or not, but I think that if we have a theory for the role that qualia serves and we build a system that appear to exhibit qualia then we could reasonably assume that it really does have an internal experience of the world akin to ours.
I see two likely possibilities:
I agree with what you say. I would call the process of working on the internal model "thinking". It would in principle be possible to have a system that thinks just like we do and appears to possess qualia, but it does not. Unless the scenario (1) is correct - and qualia will spontaneously emerge in a complex thinking system - which I think is the case. So, qualia is a necessary byproduct of our useful internal model processing.
As you have said, we can make the reasonable assumption that systems that behave as if they have qualia have qualia, then we may proceed doing useful things in this field. We may design new types of useful conscious systems, have discussions about AI and animal welfare ethics, etc.
As to:
II interpret this as if the answer of the hard problem is "we can't solve it, because it is fundamentally impossible to measure qualia". Measuring qualia is essential to address the hard problem, because if we can't show that a system has qualia, then we can't show the conditions under which it arises.
I also used to think that AI was the best way. Recently though, the paper of training neurons to play ping-pong blew my mind, because they practically showed that you can apply negative and positive stimulation to a simple network of neurons by providing it with either predictable or random information. I am not so hopeful about the hard problem being solvable, but seeing work like this gives me hope.
I think that qualia is one of the most (if not the most) interesting pieces of nature. To me it is practically super-natural. The laws of physics as we know them can describe our world without a hint of qualia, and the unieverse just randomly decided to sprinkle the gift of conscious awareness on us for no apparent reason. It is so ubiquitous (i think) and yet we can't answer the simplest questions about it with certainty.
I'm definitely partial to the idea that qualia is an emergent property, but I also think that it's a continuous gradient as opposed to a binary property. I suspect that some level of self awareness arises in any organism that interacts with its environment in a volitional way. At a fundamental level it's the organism modelling itself in its environment. Being able to move towards food and away from danger are two basic requirements, and creating an internal model that integrates sensory inputs and reacts to them is a good solution to this problem. And we observe this in pretty much all living organisms. Then it's just a matter of how complex this model becomes.
Simple organisms have a very rudimentary model, but complex ones exhibit increasing levels of fidelity. There is also a possibility that qualia is just an exhaust fume from this process. It's entirely possible that the functional survival of the organism does not necessitate consciousness, but it is an emergent property that necessarily falls out from the need of the organism to model itself within the environment. Consciousness can be sort of an echo resulting from that process.
I definitely agree that this is one of the most interesting problems in nature. It's fascinating that it's something so core to our experience, yet we're unable to define what it is exactly.