Wikipedia:Reference desk/Archives/Science/2017 June 27
Science desk | ||
---|---|---|
< June 26 | << May | June | Jul >> | June 28 > |
Welcome to the Wikipedia Science Reference Desk Archives |
---|
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages. |
June 27
editCan anyone have in their mind a representation of four or more dimensions?
editCan people (who presumably study higher dimensionality) hold in their mind a representation of those dimensions like I think I can just about hold in my mind three-dimension arrangements? ----Seans Potato Business 17:07, 27 June 2017 (UTC)
- It depends on what the meaning of "like" is. We don't have experience with four spatial dimensions, and I'm not aware of anyone who can visualize four-dimensional objects in the same way that you would visualize three.
- But there are workarounds. Take a Klein bottle, for example. It can't be embedded into three-dimensional space without self-intersection. But you can, say, think of most of it as green, and the part that comes around to self-intersect shading into red, and then back to green, whereas the part it pokes through is green. Or you can imagine coming up to the self-intersection and then looking at the area rotated, so that you see a path passing around rather than through. Or you can map one dimension to time and think of an animation.
- Five is obviously even harder. --Trovatore (talk) 17:41, 27 June 2017 (UTC)
- Six is right out. Rojomoke (talk) 22:32, 27 June 2017 (UTC)
- Some people claim to be able to do so, and that you can get better with practice. Charles Howard Hinton invented a set of colored blocks that he claimed could be used to train yourself to visualize the fourth dimension. See also: [1] CodeTalker (talk) 18:07, 27 June 2017 (UTC)
- I either use time as my fourth dimension and have a four dimensional concept be a short movie, use color, or in the case of polytopes where the majority of vertices are in a limited number of planes, just offset them with dashed connectors. (So w+1 is equivalent to x+50).Naraht (talk) 19:18, 27 June 2017 (UTC)
- Yes, using virtual reality glasses and practice. This is the second time this year this has been asked. μηδείς (talk) 19:55, 27 June 2017 (UTC)
- @Medeis: I'm not sure what that link accomplishes, aside from helping Google track who reads this page. Some archives of relevance low high high low low. This is not a complete list because the Wikipedia search results give way too little context in the blurbs. The point of most of the "low" relevance ones I mentioned is that visualizing three dimensions is not actually all that simple. The eyes have a two dimensional surface, and the rest is tricks. These tricks are not really good enough to visualize an entire block of space unless you can make some assumptions (like it's a function with one value per 2D position). Nonetheless, tricks can also visualize 4D, under limited circumstances. Wnt (talk) 20:17, 27 June 2017 (UTC)
- It is possible to be comfortable with tricks of the light, if that is all you have. — Hactar Wnt, that's nicely put. Although I would argue that we have a sense of three-dimensionality that goes beyond purely what can be represented in the retina. Still, intuition is to some extent extensible, and that's actually one of the great satisfactions of mathematics. --Trovatore (talk) 23:47, 27 June 2017 (UTC)
- Of course our sense of volume and three-dimensionality is not purely due to retinal imaging. Blind people conceive of space too. There have been plenty of blind sculptors, architects, etc. Intuition is indeed extensible, and that's well put too. WP:OR For my research, I tend to wrangle fairly fixed sets of equations specifying dynamical systems, and explore 5-20 dimensional parameter spaces at a time, visualizing with various planar slices, color coding, surfaces in space, animations, etc. Put a few color-coded 3D animations in a stack and you get to 5D slices in a fairly straightforward manner. There's lots of room up there. And while I won't claim I can picture ~5D the same as 3D, I do build up intuition and make inferences that are supported by data when I get familiar enough with a certain system. I've tried many times over the years to find real empirical research on conceptualization methods of 4-n dimensional stuff but I always come up blank :-/ SemanticMantis (talk) 00:29, 28 June 2017 (UTC)
- It is possible to be comfortable with tricks of the light, if that is all you have. — Hactar Wnt, that's nicely put. Although I would argue that we have a sense of three-dimensionality that goes beyond purely what can be represented in the retina. Still, intuition is to some extent extensible, and that's actually one of the great satisfactions of mathematics. --Trovatore (talk) 23:47, 27 June 2017 (UTC)
- @Medeis: I'm not sure what that link accomplishes, aside from helping Google track who reads this page. Some archives of relevance low high high low low. This is not a complete list because the Wikipedia search results give way too little context in the blurbs. The point of most of the "low" relevance ones I mentioned is that visualizing three dimensions is not actually all that simple. The eyes have a two dimensional surface, and the rest is tricks. These tricks are not really good enough to visualize an entire block of space unless you can make some assumptions (like it's a function with one value per 2D position). Nonetheless, tricks can also visualize 4D, under limited circumstances. Wnt (talk) 20:17, 27 June 2017 (UTC)
- Depends on what 'anyone' means. IIF you mean can some people do it then the answer is yes. If you mean can everyone do it then the answer is no. Many people aren't even able to imagine 3D objects properly, a cube is about as easy as it gets but it isn't immediately obvious to them for instance what shape a corner of a cube stuck into plasticine would look like or that there's slices though a cube producing hexagons or trapezoids or five sided figures. Dmcq (talk) 22:23, 27 June 2017 (UTC)
- 4D is possible with effort, given the number of anecdotal reports that are easily found. Given that, 5D may be possible too. Beyond that point, there is always the joke about visualising 9D by visualising nD and setting n = 9. ^_^ Double sharp (talk) 23:46, 27 June 2017 (UTC)
- Yes I don't know if anyone is able to imagine 5D well, even an image of it would be four dimensional. There's various tricks for adding extra dimensions to points like having a range of colors or time or other attributes and these can work fairly well if one doesn't want to rotate anything. Dmcq (talk) 10:38, 28 June 2017 (UTC)
- I can sort of handle 4D, but it is difficult (though it gets easier with practice). I can't do 5D at all, though; I can mentally understand how it must go on, but the mental image gets more and more difficult due to the projections of projections necessary. I imagine that if you could grasp 4D with about the same level of sureness as we all grasp 3D, 5D ought to barely be within reach, but of course that is highly uncertain. Double sharp (talk) 11:00, 28 June 2017 (UTC)
- Yes I don't know if anyone is able to imagine 5D well, even an image of it would be four dimensional. There's various tricks for adding extra dimensions to points like having a range of colors or time or other attributes and these can work fairly well if one doesn't want to rotate anything. Dmcq (talk) 10:38, 28 June 2017 (UTC)
- 4D is possible with effort, given the number of anecdotal reports that are easily found. Given that, 5D may be possible too. Beyond that point, there is always the joke about visualising 9D by visualising nD and setting n = 9. ^_^ Double sharp (talk) 23:46, 27 June 2017 (UTC)
- For additional reference, there is a lot of scientific research on data visualization of high dimensional data. It is not exactly the same as a personally poking your mind's eye up in to N-dimensional clouds, but it is a bit easier to study. See e.g this [2] scholarly paper, this [3] google AI advertisement, and this [4] quora discussion. SemanticMantis (talk) 00:32, 28 June 2017 (UTC)
- So, who are these people who can visualize how a left handed helix is going to become a right handed helix by rotating it along the axis orthogonal to it? Count Iblis (talk) 02:40, 28 June 2017 (UTC)
- How is that much harder than the left no my left picture taking thing? I can't see how you can get a pentagon from cutting a cube by imagination alone. Sagittarian Milky Way (talk) 02:48, 28 June 2017 (UTC)
- If you have a horizontal slice through a cube you get a square, If you pull up one corner of that plane so it goes past one corner of the cube you get a fifth side. It isn't a regular pentagon but it certainly is five sided. And yes you're right, a 3D helix is flat in the fourth dimension but it is still interesting to visualize. It is the difference between imagining a whole object and imagining what the object looks like from various points of view, I think of the first as feeling or internalizing and the second as seeing. Dmcq (talk) 10:11, 28 June 2017 (UTC)
- The pentagon slice seems so obvious now lol. Sagittarian Milky Way (talk) 21:21, 28 June 2017 (UTC)
- Hexagonal_crystal_family and Cubic crystal system will help you visualize how a plane cube slice can be a triangle, pentagon or hexagon Gem fr (talk) 13:53, 28 June 2017 (UTC)
- If you have a horizontal slice through a cube you get a square, If you pull up one corner of that plane so it goes past one corner of the cube you get a fifth side. It isn't a regular pentagon but it certainly is five sided. And yes you're right, a 3D helix is flat in the fourth dimension but it is still interesting to visualize. It is the difference between imagining a whole object and imagining what the object looks like from various points of view, I think of the first as feeling or internalizing and the second as seeing. Dmcq (talk) 10:11, 28 June 2017 (UTC)
- Visual system is a must read here. Don't forget it turns 2-D images into 3-D perception, though a 2-D process.
- I guess "hold in the mind a representation" means somethings like "being able to stimulates the visual cortex in a way similar to the way to would react to actually seeing a 4 or 5 (or more) dimension object". And of course it can be done, as far as mathematics are involved (and they are, pretty much):
- think of fractal objects : you can fill a plane or a cube (and even a 4-D or n-D "cube" !) with a line , in such a way that any place in the plane/cube is very close to a particular point of the line. So that point of the line is a "representation" of that place of a plane/cube
- The main obvious trouble is that the visual perception system is trained to "see" 3-D objects, neither 2-D nor 4 or 5-D objects. In my previous example, the point of the line will obviously be first interpreted as a point of a line, before a point in a plane, or a point in a cube.
- I mean, any stimulation that could be associated with a 4-D object would ALSO be associated with a 3D or a 2D image, and those would naturally take precedence over the 4-D object representation. That's where training may interfere. Obviously upon seeing a flat 2-D image on a TV screen, we naturally interpret it as 3D, while it is not. So i guess, the very same way, trained people could "see" 4D objects when exposed to real or imagined 3D things.
- Gem fr (talk) 12:55, 28 June 2017 (UTC)
- There's quite a bit involved in seeing 3D. What is on the retina is like the 2D TV screen and yet we have to associate a distance with each point and get a feeling for a 3D world from it. We see a tree through a window and we know it is much farther away and bigger than the window even though its image is within that of the window. The same has to be done to see 4D. Dmcq (talk) 15:08, 28 June 2017 (UTC)
- Speculative, unsourced, misleading, and almost absolutely certain to be incorrect. No one needs to be "trained" to perceive three-dimensional point relations in a two-dimensional projection; every person with a neurologically healthy visual cognition system can do this upon their first encounter with a TV. This is because the manner in which the photons from that projection strike the retina is (in every way that counts) identical to how a photon reflected from an actual object would strike it. Sure, there are limitations in terms of resolution/at-scale variances in wavelenths of particular photons, but we can easily imagine a hypothetical super-tv which perfectly replicates the array of photons generated off an actual object and there would be no way for the brain to even recognize that it was perceiving a facsimile.
- There is absolutely no reason to assume that the visual processing centers of the human brain could accomplish a "similar" feat with an additional dimension and every reason to assume it could not. The human brain, like the nervous system of every other organism known to have one, has evolved in the context of interpreting biophysical stimuli of a very particular sort. It's physical mechanisms for perceiving those stimuli and the various computational filters by which it contextualizes the limited data and then processes it into qualia are highly idiosyncratic, and while there have been some remarkable cases of greater or less facility with being able to draw insights through visualization, there is absolutely zero reason to assume that such abilities would extend into being able to accurately project the countless number of possible 4D images a given 3D image would be a composite "side" of. Aside from the fact that the brain lacks the appropriate filters to fill in the gaps (because literally no known organism (and thus none in our genetic legacy) has ever had to "look" upon something in four dimensions--and anything that could would strain our understanding as an organism--there's also the much more basic limitation of raw processing power; aside from needing to be organized in a radically different factor, the number of individual nodes in the neural network for perceiving the phenomena would have to be orders higher than that which exists in our comparatively "primitive" brain (even though we have one of the most involved visual field processing systems of actual, known organisms inhabiting this three-dimensional universe). There's also the fact that, for most practical applications, you'd have to be simultaneously looking at the image from a minimum of three different angles at right angles to eachother (and for actual objects, six different sides), or generating a mental image to the same effect--something which the human brain/body is manifestly incapable of doing. Sorry, but nothing about your theory makes any kind of sense. Snow let's rap 22:12, 1 July 2017 (UTC)
- Charles Howard Hinton believed he could see in the fourth dimension, it was a major avocation during his whole life. He actually made a kit that he would mail on request, to help people do the visualizing. Read his books, or Rudy Rucker's book on Hinton.63.158.87.14 (talk) 22:28, 1 July 2017 (UTC)
- Interesting--thanks for that little historical tidbit! But as his ideas are represented in our article, Hinton didn't seem to so much believe that he could visualize four dimensional objects in their entirety, as much as he had some abstract mathematical insights in to how four-dimensional bodies would intersect with three-dimensional space. Still, I'll have to remember to follow up on the Rucker's book, if only out of historical curiosity--thanks again. Snow let's rap 23:49, 1 July 2017 (UTC)
- Yep, I was right: in fact, Hinton's own essay What is the Fourth Dimension? (available here via Wikisource) clarifies that he did not believe that he could directly visualize four-dimensional structures, but rather exactly the opposite: he could only infer their existence from mathematical principles, and with pronounced limitations for intuitive visualization. Here's just a couple of numerous quotes from that article which emphasize the distinction:
"Again, to go a step higher in the domain of a conceivable existence. Suppose a being confined to a plane superficies, and throughout all the range of its experience never to have moved up or down, but simply to have kept to this one plane. Suppose, that is, some figure, such as a circle or rectangle, to be endowed with the power of perception; such a being if it moves in the plane superficies in which it is drawn, will move in a multitude of directions; but, however varied they may seem to be, these directions will all be compounded of two, at right angles to each other. By no movement so long as the plane superficies remains perfectly horizontal, will this being move in the direction we call up and down. And it is important to notice that the plane would be different to a creature confined to it, from what it is to us. We think of a plane habitually as having an upper and a lower side, because it is only by the contact of solids that we realize a plane. But a creature which had been confined to a plane during its whole existence would have no idea of there being two sides to the plane he lived in. In a plane there is simply length and breadth. If a creature in it be supposed to know of an up or down he must already have gone out of the plane.
- Yep, I was right: in fact, Hinton's own essay What is the Fourth Dimension? (available here via Wikisource) clarifies that he did not believe that he could directly visualize four-dimensional structures, but rather exactly the opposite: he could only infer their existence from mathematical principles, and with pronounced limitations for intuitive visualization. Here's just a couple of numerous quotes from that article which emphasize the distinction:
- Interesting--thanks for that little historical tidbit! But as his ideas are represented in our article, Hinton didn't seem to so much believe that he could visualize four dimensional objects in their entirety, as much as he had some abstract mathematical insights in to how four-dimensional bodies would intersect with three-dimensional space. Still, I'll have to remember to follow up on the Rucker's book, if only out of historical curiosity--thanks again. Snow let's rap 23:49, 1 July 2017 (UTC)
- Charles Howard Hinton believed he could see in the fourth dimension, it was a major avocation during his whole life. He actually made a kit that he would mail on request, to help people do the visualizing. Read his books, or Rudy Rucker's book on Hinton.63.158.87.14 (talk) 22:28, 1 July 2017 (UTC)
- There is absolutely no reason to assume that the visual processing centers of the human brain could accomplish a "similar" feat with an additional dimension and every reason to assume it could not. The human brain, like the nervous system of every other organism known to have one, has evolved in the context of interpreting biophysical stimuli of a very particular sort. It's physical mechanisms for perceiving those stimuli and the various computational filters by which it contextualizes the limited data and then processes it into qualia are highly idiosyncratic, and while there have been some remarkable cases of greater or less facility with being able to draw insights through visualization, there is absolutely zero reason to assume that such abilities would extend into being able to accurately project the countless number of possible 4D images a given 3D image would be a composite "side" of. Aside from the fact that the brain lacks the appropriate filters to fill in the gaps (because literally no known organism (and thus none in our genetic legacy) has ever had to "look" upon something in four dimensions--and anything that could would strain our understanding as an organism--there's also the much more basic limitation of raw processing power; aside from needing to be organized in a radically different factor, the number of individual nodes in the neural network for perceiving the phenomena would have to be orders higher than that which exists in our comparatively "primitive" brain (even though we have one of the most involved visual field processing systems of actual, known organisms inhabiting this three-dimensional universe). There's also the fact that, for most practical applications, you'd have to be simultaneously looking at the image from a minimum of three different angles at right angles to eachother (and for actual objects, six different sides), or generating a mental image to the same effect--something which the human brain/body is manifestly incapable of doing. Sorry, but nothing about your theory makes any kind of sense. Snow let's rap 22:12, 1 July 2017 (UTC)
Is it possible, then, that a creature so circumstanced would arrive at the notion of there being an up and down, a direction different from those to which he had been accustomed, and having nothing in common with them? Obviously nothing in the creature's circumstances would tell him of it. It could only be by a process of reasoning on his part that he could arrive at such a conception. If he were to imagine a being confined to a single straight line, he might realize that he himself could move in two directions, while the creature in a straight line could only move in one. Having made this reflection he might ask, "But why is the number of directions limited to two? Why should there not be three?"
- Still, very forward-thinking insights. And a charming presentation; I remember an almost word-for-word representation of this thought experiment presented by Carl Sagan in Cosmos which once captivated me when young and never quite let go. Snow let's rap 00:17, 2 July 2017 (UTC)
Limit of intellect
editDNFT. See talk--Jayron32 04:58, 28 June 2017 (UTC) |
---|
The following discussion has been closed. Please do not modify it. |
Is the limit of the intellect be the limit of memory?--109.252.29.219 (talk) 20:08, 27 June 2017 (UTC)
|