Wikipedia:Reference desk/Archives/Science/2017 February 16

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.

February 16

Hiding numbers in a crossed-eye stereogram.

I'm making a bunch of puzzles for an Escape Room. These puzzles need to be mildly difficult to solve - but not impossibly so. My particular need is to creatively hide a 4 digit number "in plain sight" to open a combination lock as one step in a multi-stage puzzle.

I had an idea for way to hide a number in a "crossed eye stereogram" (see Stereoscopy) like this one:

(You cross your eyes until one picture overlays the other - then it all pops out in 3D - you can try it out if you sit back a ways from your monitor).

What I'm ultimately wanting to do is to place a grid of random numbers into just the right-hand image (perhaps have them written on a chalk-board inside the photo) then to place the red pattern into the corresponding place on the chalk-board in the left-hand image.

When you 'fuse' the two images into one, I hoped that the red circles would highlight four of the numbers with the lines between showing the order they go in. When you put something in one image and not in the other, it kinda looks transparent or something. You'd only be able to figure out the combination by crossing your eyes or using a 3D viewer which I'd leave lying around someplace else.

I've never seen this trick done for hiding numbers, so to test out my idea, I made this super-simple test case:

My hope was that when you viewed the image by crossing your eyes (possibly by using a Stereoscope) to align the two pictures into one - the red pattern would indicate the four numbers in the grid that you need to open the combination lock.

What actually happens is really weird and totally not what I expected! A fuzzy region around the red pattern seems to "erase" the nearby numbers in the right-eye image!

Swapping the left and right images produces the exact same response (so it's not like the effect is caused by "left-eye dominance" or anything).

So two questions:

1. What is happening? (I'm curious!)
2. Is there anything I can do to make it work the way I'd hoped it would?

SteveBaker (talk) 04:23, 16 February 2017 (UTC)

That's your blind spot - the "scotoma" or "optic nerve disk." The actual physiology is a bit complex - suffice to say that all humans have a spot in the almost-exact-center of each eye's visual field in which there is no visual perception, and your brain just "blips" it out so you don't notice it! It is the enabler for many an optical illusion, and it's also described in the PHAK as a risk: you can't see tiny dots in the sky when you stare straight at them! In your set-up, your circles draw the viewer's eye directly to the number, causing it to fall in to their blind spot. In the PHAK, they do the exact same thing with a diagram of an airplane, demonstrating how you might not see something that's flying right toward you unless you make a concerted effort to constantly scan your visual field.

You need a "bigger" number field - the actual scotoma is tiny, so if your numbers (and circles) are even a bit larger, you won't have the problem. Or you can try to train or instruct your viewers to visually scan off-center...

Nimur (talk) 05:05, 16 February 2017 (UTC)

I don't think it's the blind spot. When I close one eye, the image the other eye is pointed at returns to normal. I suspect the culprit is that the images are too dissimilar. It's difficult to keep them aligned, and the brain may be blinding you to incongruous parts of each image. Someguy1221 (talk) 05:13, 16 February 2017 (UTC)

With Steve's setup, he's forcing the number into the blind spot by requiring you to align and focus your eye in a specific location. The instant that you close one eye, saccade occurs, and your brain and eye work together to scan over the whole visual field, moving any specific location out of your optic disk for at least part of the time - helping your brain to fill in the little blip.

Just like the diagrams in the textbook, Steve's setup is actually working against your physiology and visual perception system, forcing a specific alignment, and encouraging your eye not to scan the target-object out of its blind spot. Nimur (talk) 05:22, 16 February 2017 (UTC)

No, I just used one of those diagrams to locate my blind spot. It is not remotely close to where my eyes are focusing while overlaying the images. Someguy1221 (talk) 05:53, 16 February 2017 (UTC)

At first sight, I liked the blindspot explanation - but on reflection, I don't see how it can be true. Why should the numbers always disappear and the red diagram not? Flipping the entire image left/right doesn't change the results. On the face of it there is symmetry everywhere - so why is the red diagram clearly visible while the numbers always fade out? SteveBaker (talk) 13:32, 16 February 2017 (UTC)

• In general, your brain can't overlap disparate images very effectively - rather than seeing one combined image, you see both simultaneously. For an example, see the film Goodbye to Language, which is in 3D but sometimes uses totally different images in the left and right channels. You never see these overlap - rather, they "pop" against each other, and your attention is always drawn to sudden movements or bright/sharp objects. You're getting a similar effect here - I suspect the reason that the red circles are so visible is that they have sharp edges and clear details while the numbers just form a wall of noise, so your brain always focuses on them. Smurrayinchester 08:57, 16 February 2017 (UTC)

That does seem to be the case here. I probably need to do some more experiments. SteveBaker (talk) 13:32, 16 February 2017 (UTC)

Bear in mind that some people, myself included, can't 'see' stereograms. Have never been able to. I guess in your escape room scenario there will be a group, and most people will be fine, but still. 131.251.254.154 (talk) 10:47, 16 February 2017 (UTC)

OFF-TOPIC: Stereoblindness is a well-understood phenomenon - and if both of eyes work then it's possible to cure it. (See Stereopsis recovery). The good news for people with stereoblindness is that brain plasticity often allows people recover the capability. People who do this (See Susan R. Barry), claim to see the world in an entirely new and richer way! So I encourage you to seek out an expert.

From the escape room perspective, stereoblindness affects about 5% of the population and our problems are designed to be solved by a team. The smallest team allowed is usually 4 and the largest is 8. The odds of four of them all being stereoblind is vanishingly small (5% x 5% x 5% x 5% = 0.0006%). We don't shy away from problems that colorblind people can't solve for the same reason - and a reasonably good IQ is needed to be successful too! SteveBaker (talk) 13:20, 16 February 2017 (UTC)

Perhaps not what you have in mind, but personally I find it much easier to combine overlapping images by putting one on top of the other and holding them up to a strong light. Of course, then you wouldn't be using the stereoscope, etc. I wonder if a flip book or zoetrope would be any more effective? Though again that would require a different approach. Dragons flight (talk) 11:09, 16 February 2017 (UTC)

Note that the image of the polar bear is made for parallel-eye viewing, not crossed-eye viewing. If you try it with crossed-eye viewing, the depth is reversed. And I never managed parallel-eye viewing without a stereoscope, although I've no difficulty with crossed-eye.

If, in the image with the yellow squares, I look with my left eye at the number field or with my right eye at the red lines, from a reasonable distance from my monitor, the other feature falls into my blind spot. It's far enough from the centre of my field of view not to interfere with the overlapping images. But strong features, like the red lines on yellow background, override weak features, like a wall of numbers. Furthermore, the images have too little in common. When I don't focus on the edge between the yellow and the grey areas, I can't keep the images aligned. On the other hand, I often see those "Find the 10 differences"-images on yoghurt bricks and the like, and I can easily solve them by overlapping both images. I can see the differences blink whilst the rest of the image is static.

If you want to do something with stereograms, what about hiding the numbers in the third dimension of an autostereogram? I prefer them with crossed-eye viewing (so that's the opposite of most examples in that article), but even if you view them the wrong way, the depth is just reversed but the numbers should still be readable. PiusImpavidus (talk) 11:38, 16 February 2017 (UTC)

I had already planned to use an autostereogram to provide an instruction for another puzzle. A decent escape room needs maybe ~30 puzzles to solve in order to keep a group of 4 to 8 people busy for an hour - so there is room for all kinds of sneakiness. The thing about autostereograms is that people would completely expect to have to view them to get the answer - this one would be a little more subtle (if it actually worked!).

I never understand the difference between "parallel viewing" and "crossed-eye viewing" - it seems that what I have to do is to relax my gaze and let my eyes cross in order to see the polar bear...and when I do that, it seems to have the correct depth information.

What started me off on this is that McDonalds are currently giving away a Batman-themed stereogram viewer with their "Happy Meals" and one of them got abandoned at my house by the grandkids. Since my escape room is themed as "Escape from toyland" having a children's viewer present seemed like an opportunity too good to miss! The polar-bear image works in their viewer - so I guess that "parallel viewing" is what I mean here.

SteveBaker (talk) 13:20, 16 February 2017 (UTC)

"Crossed eye" means that a line drawn out from the pupil of each eye would cross, while parallel means they go off to infinity. "Relaxing" your gaze makes your eyes parallel. Guess: The catch I think is that some people have eyes further apart then others, and the ones with their eyes furthest apart find existing parallel eye stereograms easiest to read, so they make more... for people like them. The ones who are narrower aren't programmed to have their eyes point apart in two different directions, so they never have a chance. Wnt (talk) 14:40, 16 February 2017 (UTC)

• I believe you have encountered the effect known as binocular rivalry. It is well known in perceptual psychology and has been studied quite extensively. It has nothing to do with the blind spot. Looie496 (talk) 14:09, 16 February 2017 (UTC)

I agree - as the article describes, I saw the numbers blotted out by the shape, then perfectly visible with the shape also. The catch though is that the numbers don't seem to blot out the red lines so intensely, so it is entirely possible for me to find the circled numbers - IF there are enough cues to line the shapes up accurately. Also, there is a sort of "lasso area" around the red line figure that blots out the numbers, but further away there is no interference, and the way in which that area is delineated is likely to be intensely interesting. This could be fun to look up further. Wnt (talk) 14:27, 16 February 2017 (UTC)

Fair gambit! I can concede that this may be a purely psychoperceptual illusion, like binocular rivalry, and that may even be a better explanation for the effect than positing a physiological cause like the optical disk blind spot - besides, I can't see what you see, so maybe we're seeing (and explaining) different effects! This raises some fun questions about testability and falsifiability as it pertains to visual illusions: how could we test the root cause? If it's your blind spot, scanning the field should make it go away (and does, for me at least). I'm not certain that proves very much, though!

At the very least, psychological and perceptual illusions can coexist with real physiological visual artifacts - it could be a combination of all of the above!

Nimur (talk) 06:57, 17 February 2017 (UTC)

@Nimur: I don't understand how you can look at that image and think the blind spot is anywhere near what you see. I mean, the images in each eye that don't superimpose are somewhere near the blind spot, but I can remove them from play simply by holding up my hands as a frame so that those images are blocked out, and the superimposed parts still look exactly the same as they did. Wnt (talk) 20:53, 18 February 2017 (UTC)

I'd like to suggest that you use a rather complex image for the stereograph, but then make the hidden message different in the two eyes. When looking at a stereograph, any details that don't "match" stick out like a sore thumb. You notice this when using vintage stereo cards that have a scratch or something on them. It feels like the scratch is flashing as your two eyes fight over what version of the image is 'correct'. (You can intentionally exploit this effect to quickly solve one of those 'spot the differences' puzzles they sometimes print in newspaper comic pages.) ApLundell (talk) 17:27, 16 February 2017 (UTC)

 ASCII-art stereogram:

 Here is an ASCII stereograms.  View with a monospace font.
 To see the 3d effect, you need to focus your eyes such that the two
 adjacent "X" characters at the top look like three. Once you've done 
 that, wait a few moments for your eyes to get used to it, then slowly 
 look down at the rest of the image.

                             X         X
 .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .
  .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .
    .    .    .    .    .    .    .    .    .    .    .    .    .
    .     .     .     .     .     .     .     .     .     .     .
      .      .      .      .      .      .      .      .      .
  .       .       .       .       .       .       .       .       .
       .        .        .        .        .        .        .
    .         .         .         .         .         .         .
 .          .          .          .          .          .          .
 |          |          |          |          |          |          |
 |          |          |          |          |          |          |
 |          |          |          |          |          |          |
 |          |          |          |          |          |          |
 |          |          |          |          |          |          |
 |          |          |          |          |          |          |
 .          .          .          .          .          .          .
    .         .         .         .         .         .         .
       .        .        .        .        .        .        .
  .       .       .       .       .       .       .       .       .
      .      .      .      .      .      .      .      .      .
    .     .     .     .     .     .     .     .     .     .     .
    .    .    .    .    .    .    .    .    .    .    .    .    .
  .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .
 .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .

--Guy Macon (talk) 19:06, 16 February 2017 (UTC)

When I look at the image with yellow background, I get the stereo effect, but the red circle's pattern does not "lock" on the random numbers. Circle's pattern kind of vibrates in horizontal position. Even then, I can see that the area of red circles overwrites the background numbers. But I noticed, with some efforts, I can see the numbers but this make the circle kind of disappear. I was able to focus on either on circle or on number at a time, and was able switch focus as desired. Therefore, to get images in locked state, I thought I need a matching pattern in remaining part of both images. Then actually I cut the red circle's pattern and pasted on the right side of bear on the left bear image, and random number pattern on similar position on the right bear image. And then when I looked, I got the circle pattern nicely locked on the numbers. And in stereo view, I could still focus on either circle and on number, and flipping the focus revels the desired numbers! Give it a try. manya (talk) 05:36, 17 February 2017 (UTC)

If you don't mind it not being a proper stereogram (like the one with the polar bear), there's something called SIRDS (single image random dot stereogram.) There are generators for that kind of thing on the internet if you google. They take an image that contains a mask and create a "Magic Eye"-type image that contains the outline of what was in the original image. When viewed properly it will appear embossed Asmrulz (talk) 17:06, 17 February 2017 (UTC)

Here's a random one from google search results: http://pictureimage.whak.com/signs/stereogram/?text=123

It takes a string of text and generates a SIRDS image with the string hidden in it. I personally couldn't "acquire lock" on the image but I think it's because I'm viewing it on a monitor. It'll work if you print it out (perhaps scale up a bit) Asmrulz (talk) 17:14, 17 February 2017 (UTC)

Identify a tadpole shrimp (Triops)

Is this photo a Triops granarius or a different species? Found in an ephemeral pond near Victoria West, Northern Cape province, South Africa. The image in the article is too small and unclear to properly compare it. Roger (Dodger67) (talk) 07:42, 16 February 2017 (UTC)

Hmmm, a quick search online yields an essay suggesting that there is some work needed on the taxonomy. [1] I thought someone must have looked at the DNA, so I went to PubMed and found this, which should be useful. But I'm a bit reluctant about using Sci-Hub from this location, and who the heck has access to "Zootaxa"? So I'll leave it to you to go over what they say about the two other species they think they can distinguish there. Wnt (talk) 14:35, 16 February 2017 (UTC)

Thanks Wnt, leaving the taxonomic fights aside, it seems that granarius (by whatever name) is the common species in the area concerned. Roger (Dodger67) (talk) 16:59, 16 February 2017 (UTC)

does exercise oxidize us

does exercise oxidize us? Is this unhealthy somehow? — Preceding unsigned comment added by 31.4.145.73 (talk) 15:42, 16 February 2017 (UTC)

This may be useful for your research. --Jayron32 15:47, 16 February 2017 (UTC)

In the literal sense, yes, because the food you eat is oxidized to release energy for your body to use. As with anything, extremely excessive physical activity can be bad for your health, but the health benefits of reasonable amounts of exercise are well-documented. If you live in the developed world, you're at much greater risk of not getting enough physical activity, which can lead to all kinds of health issues. See also oxidative stress for the general topic of cell damage caused by oxidation reactions. --47.138.163.230 (talk) 01:56, 17 February 2017 (UTC)

We can read a lot about this subject in scientific articles and our own articles as mentioned above. But we need to consider here that our understanding of the processes involved here is still quite new while the biological processes are very ancient, many of these date back to the time that organisms first started to use oxygen about 2 billion years ago. Evolution has led animals that have made the best out of all these processes, so a naive way of thinking that heavy exercise may not be healthy due to the large amount of oxidation, probably does not reflect reality. The opposite, taking a lot of anti-oxidants has been put to rigorous tests, with typically negative outcomes, e.g. taking a lot of vitamin C impairs recuperation after heavy exercise. Count Iblis (talk) 02:26, 17 February 2017 (UTC)

for your last claim, preferably with a WP:Medrs compliant source if you can. It's not mention by the ref linked by Jayron32 above. Nor in vitamin C which only mentions "A study in rats and humans suggested that adding Vitamin C supplements to an exercise training program lowered the expected effect of training on VO2 Max. Although the results in humans were not statistically significant, this study is often cited as evidence that high doses of Vitamin C have an adverse effect on exercise performance. In rats, it was shown that the additional Vitamin C resulted in lowered mitochondria production.[40]". I assume we're talking about humans not rats since the OP clearly is and in any case I don't think this single study really belongs in our article (but that's something for the article talk page). And "It has not been shown effective in prevention or treatment of the common cold, except in limited circumstances (specifically, individuals exercising vigorously in cold environments).[25][needs update][26]".

I'm not suggesting that vitamin C supplementation after or before heavy exercise has any benefit. Rather most likely there is no benefit nor negative, e.g. as mostly suggested by this (non medrs) source [2], e.g. "The overwhelming consensus of a fairly large number of well-conducted investigations is that vitamin C has no ergogenic effect in persons who are not vitamin C deficient" and "The overwhelming consensus of the literature is that long- or short-term supplementation with vitamins E or C has no ergogenic effect on submaximal exercise performance, aerobic capacity, or muscle strength". Yes this only talks about ergogenic effect but it's silly to think they wouldn't have mentioned if there was a well supported ergolytic effect.

Note this doesn't mean it's a good idea to supplement as some people claim because it probably has no significant negative effect. Since you have no reason to think it has a positive effect either nor to think a positive effect is more likely, that doesn't make sense. And there is the minor (presuming you aren't talking about extreme amounts) stress you're putting on your body excreting all that excess vitamin C.

Nil Einne (talk) 06:28, 17 February 2017 (UTC)

Thanks, I'll try to find the articles that I saw some time ago, but it may be that these are the same as you quote here. Count Iblis (talk) 18:16, 19 February 2017 (UTC)

How do I make a factory?

I really like watching How It's Made series on YouTube. Everything is automated. There are machines that control everything in high speed! I wish I can make my own factory. What skills and knowledge and how much money do I need to build my own factory? How many subjects in physics and chemistry do I have to learn about before I can make my own factory? How do I extract metal ore from the ground? How do I convert metal ore into usable pieces of metal? How do I make a wire? Where do I start in the process of making an automated machine? 66.213.29.17 (talk) 19:51, 16 February 2017 (UTC)

Are you the same person who asked the question about factories a few days ago? If so, the answers are there still. To reiterate: One person does not make a factory. Hundreds do. It is beyond the scope of ability and knowledge and time and resources for a single person to handle the entire thing. Please scan up this page for responses to this question the last time it was asked. If there are more specific questions that you need answered, we'll try to help as well. --Jayron32 20:01, 16 February 2017 (UTC)

#What type of person makes all the factory parts and programs them to work?

For the historical aspects, some good reads are John G. Landels'. Engineering In the Ancient World. and L T C Rolt's. Tools for the Job.Andy Dingley (talk) 20:24, 16 February 2017 (UTC)

We have articles on mining, ore and metalworking. I recommend typing some of your keywords in to our serach box, and then reading the articles. We can help you read up on these things, and answer questions about them, but we can't teach you the entire field of industrial engineering and design of factories. SemanticMantis (talk) 21:52, 16 February 2017 (UTC)

"How do I make a factory?" is not a science question. It should be asked on another page. DrChrissy ^(talk) 21:58, 16 February 2017 (UTC)

The same IP user posted this question about a factory. Blooteuth (talk) 00:44, 17 February 2017 (UTC)

For a factory on the human scale I might recommend RepRap and other rapid prototyping and 3D printing. For a chemistry factory, microfluidics is a growing field that carries out whole reactions in a tiny module. While these technologies currently have significant limitations relative to conventional methods, they seem to have great potential, and they allow new methods - for example diazomethane can be produced and used in a microfluidic set-up, even though a traditional chemist might not dare to come within sight of a flask of it. Factories have largely shed their human workers for machines; arguably the next step is to make those machines tiny, and the factories to match. Wnt (talk) 12:46, 17 February 2017 (UTC)