This sheds some light on one of those things I believe, but can not prove...
I usually carry too much junk in my pockets, and when I dig around in my pocket for something it amazes me how fast I can recognize what I'm touching. I don't know why but I've thought that there is some other non-conscious system in place that, at a minimum, helps improve tactile recognition. It's kind of subtle, but it feels like my hand somehow knows what I'm searching for, and can recognize it much better than I can.
Also, along the same lines of thought, I suspect that there must be a whole lot of intelligence locked away in our bodies that isn't easily accessed consciously, but it's still there.
I think this concept has existed for quite some time. Musicians and athletes and dancers have always known about muscle memory. But whether those memories existed in the appendages or in the brain was never well known. I think this gives more credit to the idea that some memories are actually stored in the Peripheral Nervous System.
As a thought experiment, lets say a famous guitarist loses an arm and gets a perfect transplant or a prosthetic which perfectly matches their previous dexterity. Would it perform as well as their old hand, or was the hand performing additional calculations on incoming and outgoing data that could never fully be replicated by a replacement.
Now take it a step further and assume a random person receives a famous guitarists arm in a perfect transplant. That arm is an instrument in itself with its stored knowledge, but that doesn't mean the receiving brain knows how to communicate effectively with it. I'm sure they would have basic communication, but the extended neural vocabulary of the old Brain-hand pair would be known to the new host-brain.
Personally, I believe it is a tandem effort between CNS and PNS in which each becomes accustomed to the signals sent by the other.
For example: A brain tells a finger to curl HARD, the finger neurons receive the brains input, modulates it with local factors, and sends it to the muscle.
Likewise, the finger knows the feel of the strings exceptionally well, and can send highly detailed messages back to the brain who is used to receiving highly detailed messages. But in the new host who doesn't know this advanced feeling, it just feels like a regular old input.
> It's kind of subtle, but it feels like my hand somehow knows what I'm searching for, and can recognize it much better than I can.
Most brain processes are subconscious, so if neurons outside the brain perform part of those processes, it should not really matter. Subconscious processes are subconscious, within the brain or not.
Anyways, read "On Intelligence" by Jeff Hawkins for a theory of how very fast pattern matching works in the brain. Skin neurons not needed, but can also fit into the theory I think.
Don't want to sound harsh, but have you really always thought like that, or just now that you read the article? I am just asking because such things occur to me often: I read something I always had a vague idea about, then read it in good wordings and suddenly I'm like 'oh yeah, I've always known that'. While I obviously haven't. Think carefully before giving an answer, the brain is king into tricking 'you' into believing things.
it feels like my hand somehow knows what I'm searching for, and can recognize it much better than I can
Imo that is just your brain having learnt your habits in such a way that you now don't actively have to think about it anymore. The sensory inputs of your hand still go to your brain, and are processed there into commands to your hand's muscles. Subconscious or not, sensory input locally 'pre'processed or not, your hand is not locally processing that complete cycle so it doesn't 'know' anything nor is it working standalone.
This kind of result, of which there have been many, is pretty of the reason why I believe the computing power needed to rival human intelligence is vastly underestimated. And why singularity predictions are fundamentally overoptimistic.
The other major reason is that much of cognition happens outside the body, in physical and cultural structures that are even less computationally tractable.
> This kind of result, of which there have been many, is pretty of the reason why I believe the computing power needed to rival human intelligence is vastly underestimated. And why singularity predictions are fundamentally overoptimistic.
I think you should look at the numbers some more before issuing confident proclamations like that: the brain has something like 80 billion neurons. The famous gut has more like 0.1 billion neurons, the heart has more like 0.0004 billion neurons, and one's skin isn't going to contribute that much. The number of neurons outside the brain is much much less than the uncertainty in current estimates of how many neurons are in the brain - they're just not an issue. If one can compute something approaching the brain, then one can certainly handle the rest with a spare 1% of capacity or so.
Likewise, a great deal of what we find 'important' or 'significant' relates directly to our basic existence as material bodies with specific biological needs. Like a fish not realizing they were in water, it is very easy for people to not notice how reliant they are on their bodies. Were we able to build a machine intelligence, I have serious doubt that we would be capable of recognizing it as alive or conscious. It would have absolutely no drives in common with us. It would not need space. It would not need the separation of individuals. It would not need to respond to stimulus on a short time scale. It would probably have absolutely no need to communicate with us at all. Its needs would be so utterly alien to us that it would be very difficult to observe it and conclude 'that is intelligence'.
Likewise I think the idea of "uploading your brain" is ludicrous unless the plan is to simulate it existing in a virtual physical world. The human brain reacts poorly to sensory deprivation - consciousness dissolves in the absence of physical sensory input. Cutting off every sense totally? I don't think we have any evidence to suggest that what would remain could qualify as a human mind. And we've got some evidence that it would not. Our bodies and our brains are not separate. Dualism is fundamentally untrue. We can see this in many studies of various injuries that rob people of various capabilities of their body which bleed over to significant effects on their mind. Make people hold a pencil in their teeth, which forces them to 'smile', and they will have more positive emotional reactions. People with paralyzed facial muscles suffer decreased emotional range in their mind. Paralyzed individuals generally experience depression (separate from the depression expected from suffering harm). Cutting off the brain from the body hurts the mind.
This isn't surprising or really new. There are 4 types of mechanoreceptors in the skin, plus pain and temperature sensors. Each of those types have individual endings or sensors that have a specific range of intensity to which they respond most strongly. Also, they compete locally with each other. So the idea that neurons are "computing shapes" is really no different than computing edges, corners, specific patterns of vibration, a rapid change in temperature, or any other sensation that is more than a single point in time and space. In general the peripheral nervous system does a lot of computation and, given that many species don't have, or can do without, a brain, we shouldn't be surprised when things that are done in the neo-cortex are also done by the antecedent computational substrate.
I'm thinking about the findings in skin neurons from a slightly different angle, well maybe two angles.
For one, study of brain increasingly points to widely distributed circuitry as the way functionality is implemented. But the brain doesn't exist isolated from the rest of the body, on the contrary, endocrine, immune, gut systems are highly interconnected with brain circuits.
IOW it's plausible neural circuits throughout the body are part of the brain's distributed circuitry, and contribute to functions we identify as "brain functions". The idea of "brain" as synonymous with "brain function" may be a great oversimplification.
The second idea might support a concept of body-wide distribution of information processing circuits. In embryogenesis, brain and skin both stem from the outer layer or ectoderm of early embryonic development.
We can speculate this common origin means neurons in the skin, and throughout the nervous system, are endowed with computational capability and continue to participate in non-local computations. This idea implies neurons winding up at locations far from the brain could still be included in the distributed network.
It will be interesting to find out where the research on neurons in fingers takes us. It might turn out to be a wild ride.
We do know that amputations and organ transplantations are possible without causing brain damage, though. So this can only be true to an extend - or at least the body has the ability to tolerate loss of information.
The problem in Parkinson's is degeneration of midbrain dopamine neurons which project to other brain regions that regulate and coordinate body movements. It's hard to imagine what connection would exist between neurons in the skin and the dopamine neurons in the brain.
That's not to say the skin's neurons aren't amazing in their own right. Very interesting research indeed.
this makes perfect sense after reading some molecular biology lately: bacteria are constantly trying to attach and detach from our skin, and that, apparently, involves a constant struggle between electrostatics and biological structures...
... my issue is whether or not that ought to be called "calculation" or "computation." sure it's an /interpretation/ of the natural process, but not necessarily equivalent. I realize that's not what this article is saying, but i wonder if we're getting too bogged down in algorithms with some of these things.
_in_Heteractis_magnifica_(Sea_anemone).jpg){kind=link}
I usually carry too much junk in my pockets, and when I dig around in my pocket for something it amazes me how fast I can recognize what I'm touching. I don't know why but I've thought that there is some other non-conscious system in place that, at a minimum, helps improve tactile recognition. It's kind of subtle, but it feels like my hand somehow knows what I'm searching for, and can recognize it much better than I can.
Also, along the same lines of thought, I suspect that there must be a whole lot of intelligence locked away in our bodies that isn't easily accessed consciously, but it's still there.