4 ms·
But is there infinite regress -- does it have a representation of itself representing itself. I find a more prosaic version of this question quite interesting
by srean 5y ago
But is there infinite regress -- does it have a representation of itself representing itself.
I find a more prosaic version of this question quite interesting too. If you can feel a touch/heat anywhere on or in your body its represented in the brain somehow, somewhere. For a moment (i) imagine the brain is sensitive to touch (it isn't because lack of 'pain' receptors) (ii) that the representation of a sensitive part of our body is localized in the brain rather than distributed.
What this gives us is a mapping from the entire body (that includes the brain) into itself. Now, if there is a fixed point of this mapping, that fixed point is aware of itself.
Is the brain insensitive to touch because its not biologically/physically possible for a localized region of space not to be aware of itself ?
EDIT: Downvote ! That was rather unexpected.
- skohan 5y agoYeah sure it does - probably you can conjure a cartoon in your mind of a brain modeling a brain modeling a brain all the way out to infinity. Maybe you already did when you were coming up with this problem. edit: since the other poster edited their comment, this was only responding to the infinite regression question, not their bizarre point about touch sensitivity
- XorNot 5y agoSimilar to the idea of the "infinite image". Take a 640x480 square, and permute every possible combination of pixels. By definition: that space must contain a sizeable, or even complete, portion of all human knowledge that can ever be scrolled or exist in that image space. A lot of noise too, but the complete cure for cancer (if possible), FTL (if possible), immortality: what answers absolutely cannot be represented as several variations of the pixels of a 640x480 image?
- skohan 5y agoTo respond to your edit, I don't quite understand what you're getting at. It's quite well understood how the brain represents touch - all the sensory neurons in your skin send information to a brain region which maintains a spacial representation of your body. Your brain isn't sensitive to touch because it's inside your skull, so that touch information wouldn't do a whole lot for you. There's no reason in principal a brain couldn't have pressure sensitive neurons inside of itself or on its surface.
- srean 5y agoNo. Even you crack the skull open and poke the brain, it wont feel a thing even when conscious. The skull, the skin will feel, but the brain itself has no pain/tactile receptors. Many surgical procedures on the brain are done with the patient very much awake.
- skohan 5y agoYeah exactly, why would it?
- srean 5y agoI am just adding to your explanation. The brain is insensitive not because it is "inside your skull" but because of the lack of pain receptors in the brain. Our bone marrow is inside our bones. Trust me, it is not pleasant to touch.
- skohan 5y agoYeah that's exactly what I was saying. It doesn't need touch receptors because it's inside your skull. It makes sense to evolve pain sensitivity in your bones because if you are doing something which is causing damage inside your bones, you had better have a signal to stop. For most of the evolutionary history of humanity, if your skull was compromised, you would be dead.
- throwanem 5y agoI mean, the brain isn't sensitive to touch because it doesn't contain the kind of mechanoreceptors that our nervous system uses to detect that kind of stimulus, which seems like a perfectly sufficient explanation here.
- srean 5y agoIndeed. What I am asking is something different, although I did not word it properly -- Is it even possible for the brain to be sensitive to touch everywhere, or must a teeny-weeny dead-zone exist.
- skohan 5y agoWhy would it need to exist? I feel like you're reaching for a philosophical argument here for a question which is already adequately resolved by physics and biology.
- throwanem 5y agoIt seems like the question is fundamentally that of whether a given system can perfectly model its own state, but in this context that entails an implicit assumption that the brain's model of body state is perfect, which it is anything but.
- skohan 5y agoI guess I don't really think any system would be able to model it's own state 100% accurately. Basically any model implies a reduction in information. I don't see what this has to do with pain perception in the brain. I am sure you could find an example in the animal kingdom of a creature which has pain receptors in its brain if you look hard enough.
- srean 5y agoDon't get hung up on 'pain perception', my point sits at a higher level of abstraction. It has very little to do with biology, more to do with topology. If 'events at point_a' gets consciously interpreted by participation of cells one of which, lets call it point_b, is singularly dedicated to sensing events at point_a, we have a mapping in a mathematical sense from point a to point b. We know all such point_bs form a strict subset of the body, a part of the central nervous system. If this mapping is continuous in the mathematical sense and a contraction, then a fixed point has to exist (a location that does its own interpretation). But the fixed point seems biologically implausible, that leaves us with the alternative that there has to be dead-zones -- places where events can happen but cannot be interpreted. (Example blind spot in the eye) The way to avoid dead-zone would be to break continuity, or contractive property of the mapping. > I am sure you could find an example … of a creature which has pain receptors in the brain … I think my point would be easier to state for this hypothetical creature. If interpretation of pain is done by a strict subset of its cells and there is a one to one mapping -- at least one location b that interprets pain occurring at location a exclusively (presumably that's how the creature knows the pain is at 'a'), can this creature be free of dead-zones if the mapping is continuous and contractive. Can any creature free of dead-zones exist under those conditions. You claimed 'why not'. I am saying that if the mapping is a continuous contraction then either there has to exist dead zones or there is at least one self interpreting location. To give another example of this phenomenon, if you spread the map of your city somewhere in your city, there has to exist at least one point in the map that sits exactly above the point it is representing -- the fixed point. The way to avoid this would be to have 'tears' in the map, or to have points in your city that are not closer on the map than they are in the city (that would be an unusual map).