On Sep 1, 3:18 pm, meekerdb <meeke...@verizon.net> wrote: > On 9/1/2011 9:21 AM, Craig Weinberg wrote: > > > Whether or not afferent and efferent nerves are fundamentally > > different kinds of cells or just playing different roles in the > > nervous system isn't important. If they are the same that only makes a > > stronger case for me, since there would then be no biochemical > > determination of that role. > > But there would be a structural determination their role; one depending on > what other > cells they were connected with. On the neural network computer model of the > brain it is > these connections and their strength that are analogous to the hardware and > the software.
To explain why a neural network would arise organically though, you would have to cite some way for the larger network structure (which it sounds like you are saying is an abstract, metaphysical construct made of 'connections') to arise independently of the component cells and determine their roles for them. The abstraction of 'calculation' is the active agent, while the cells are interchangeable passive variables which would execute the calculation just as well whether it were made of bowling balls or ice cream sandwiches. > When the brain is active, it is processing information But that's not all it's doing, It's also experiencing life as a human being, which is not found in any 'information' being computed. > and at the abstract level of > computation the neural network computer performs the same computation and is > functionally > equivalent at it's input/output. I'm saying that computation can only emulate 50% of the input/output. The other 50% arises from private sensorimotive perception and is not computable, but rather it chooses how to interpret some of the input and decides what to output. Input/Output supervenes on this middle layer of private interpretation, which, I think is what is enhanced through qualitative morphological synergies (molecules>cells>organisms>brains...). A molecule has a relatively narrow middle layer, a cell has more degrees of freedom (more qualia to feel and more 'time' to feel it in, more time and teleological depth of field to project 'will') than a molecule, an organism has more than a cell, a brain has more than an organism with no brain, etc. > Whether it instantiates the same qualia (or any at all) > is a different question. Since we know that whatever functions can be > computed by finite > things (including brains and computers) can be computed in this way, we infer > that it is > possible to make an artificial brain that will produce the exact same > responses as a real > brain and in a suitable robot/android will produce human like behavior. Of > course the > question of its qualia (if any) remains open. I understand why it's compelling to assume that it is possible to make an artificial brain that would emulate a natural brain, but it's based on a critically flawed model of what the brain is. Consider the brain to be the physical shadow of the self. Duplicating the shadow of a palm tree, is not going to suddenly cause a palm tree to be conjured into existence. Yes, the shadow is the same exact shape but it doesn't behave the same way, even though it seems like it should since we've meticulously modeled the way the shadow sways with elaborate algorithms. You assure me that we have studied every possible change to the shadow and it should therefore produce the exact same responses as the real shadow. The problem is that the shadow is not what is determining the motion. It's the palm tree in it's environment of wind, rain, birds, etc. Same thing with the brain, although it's bi-directional. We are influenced by our brain and vice versa, so we are each other's shadow side. Our feelings and experiences alone determine how our brain will behave in part, and that's the part that cannot be emulated by computation without being able to live our life from our subjective point of view. The other functions that the brain is doing could maybe be deduced from the status of the rest of the body and emulated with greater success, but that too may be a naive reduction of the brain's relationship with the body. That's not to say you can't make a recording of the brain activity of someone and have some success playing it back on another - just as a computer playing an mp3 of a Mozart symphony doesn't need to know how to emulate the activity of a philharmonic orchestra, but they are two very different things. The computer is not it's own user. The brain is. > > > You're not answering my questions though. > > I know we have different feelings about different things, I'm asking > > you why anything has any feelings at all, and is a feeling a physical > > thing or not? > > There are different ways of answering a "why?" question. In this case, one > answer is that > feeling is a physiological response to the environment. We have such > responses because > they are, or were, advantageous in survival and reproduction and hence > selected in the > evolutionary process. This explains why we have lots of tacticle sensors on > our surface, > where we can react to things, and not so many in our digestive tract where > our responses > are limited. Feelings are physical, but they are not things (i.e. objects) > they are > changes in things, e.g. hormones released into the blood stream. That doesn't explain anything. Again, our stomach digests things, our immune system handles much more complex and important tasks related to our survival and reproduction without our feeling anything. There is no mechanical advantage, nor is there *any* possibility that feeling can arise from physical evolution. As you say, they are not objects, so they cannot evolve. 'Changes' in hormones in a blood stream don't just decide that they are a 'feeling'. There is no 'they'. A change isn't a thing that feels. It's only the cells themselves, or the tissues they make up that could possibly feel these changes. The appeal to 'changes' and 'responses' as sense agents is metaphysical. > But you are presumably asking about feelings as emotions: thoughts of joy or > sadness or > satisfaction or anxiety. No, you had it right before. I'm asking about sensation and the interpretation of sensation (which I call feeling). Interpretations of what I call feelings are what I would call emotions, and the interpretation of emotions are what I call thoughts. >These are the same physiological changes sensed at the level of > consciousness in humans and put into an inner narrative. Evidence for this > is the fact > that various drugs can produce these feelings independent of other changes in > the environment. There is a physiological side of emotion and an experiential side, just like everything else. You can manipulate emotions physiologically, and you can manipulate physiology emotionally. It's bi-directional: bottom up AND top down. > What are the implications of this for the robot/android whose artificial > neural network > brain produces human like behavior? Well it's obvious that a silicon based > brain won't > respond to LSD or oxycontin like a human one. And similarly the silicon > brain will > respond to an EMP that an real brain won't even notice. Right, which is why we know from the start that a silicon brain can never emulate ALL of the behaviors of a natural brain. > The question then is whether this > shows the artificial and human brains instantiate different qualia even when > their > behavior is the same or only when there are these different responses to the > environment. It's not that the inability to respond in the same way as a natural brain equals different qualia, but it certainly should be an indicator that it very well could, especially if we think that qualia is related to electrochemical processes. Mainly we have no reason to imagine that a silicon brain has any other qualia beyond that inherent in it's physical manufacture and operation. It doesn't learn to feel like a person, just as your computer doesn't learn how to see you through the monitor. It's not alive. It has as much qualia as a shoe. Craig -- You received this message because you are subscribed to the Google Groups "Everything List" group. To post to this group, send email to everything-list@googlegroups.com. To unsubscribe from this group, send email to everything-list+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/everything-list?hl=en.
