Wikipedia:Reference desk/Archives/Science/2018 May 17

= May 17 =

Leaving group vs Nucleophile
This question relates to nucleophilic substitution and elimination reactions in organic chemistry. If a particular species is a good nucleophile in a given solvent, does that then make it a poor leaving group? One of the characteristics that makes a species a good nucleophile is that it's poorly solvated, thus freeing up its surface and movement for interaction with a haloalkane. Conversely, one of the factors that makes a good leaving group is that it's well solvated, and leaving the haloalkane to form a carbocation is energetically favorable (or at least not very unfavorable). Given both of these facts, it stands to reason to me that if a given nucleophile is poor in a given solvent, then it should be a good leaving group in the same solvent. Further, if leaving group A is displaced by nucleophile B, but A is also a good nucleophile, then it should simply go back and displace B in an endless cycle. Of course an equilibrium of such alternating displacements is established in real world reactions, but the equilibrium position lies far to one side or the other in useful reactions. Despite all this, it seems to be generally accepted for example, that Br is a good leaving group, and Br- is a good nucleophile, and this is often stated without specifying the solvent (which makes it a bit meaningless to me). My text book includes a table of nucleophiles and their strengths in water/ethanol solution. It places Br- as stronger than Cl-. In a question in the chapter it asks a question that essentially boils down to which of the two halides is a better leaving group in ethanol. It says the answer is Br- which seems inexplicable given they already said it is a stronger nucleophile in almost exactly the same medium. 202.155.85.18 (talk) 10:03, 17 May 2018 (UTC)
 * You're generally right that good leaving groups make poor nucleophiles, and visa-versa, but there are some exceptions, and the halogens are often exceptions to these rules. For example, see Electrophilic aromatic directing groups, in electrophilic aromatic substitution, halogens are ortho-para directing, but are electron withdrawing groups; all other electron withdrawing groups are meta-directing; and all other ortho-para directing groups are electron donating.  The reasoning for this may be closely related to the reason why halogens also buck the trend in nucleophilic substitution trends.  -- Jayron 32 11:16, 17 May 2018 (UTC)
 * Thanks for that response. In that case, if a halogen like Br can be considered both a better leaving group and a better nucleophile in a polar, protic solvent such as water or ethanol, then what do we expect to happen when say sodium bromide is added to a chloroalkane in ethanol? Do we expect the Br to displace chlorine because it's a better nucleophile? If so, does it then immediately form a carbocation because it's also a better leaving group? While I appreciate that the halogens have some unique behaviours, at some point these concepts of good nucleophiles and good leaving groups become mutually inconsistent. 114.124.243.188 (talk) 23:13, 17 May 2018 (UTC)
 * And that's why organic chemistry is hard as balls. Real chemistry is messy as heck, because you've got dozens of competing principles, and trying to quantify how each one applies in a particular reaction will make you pull your hair out.  This page actually has a good discussion of the solvent effects on nucleophilicity of halides.  I think it can help you work out the answers to your questions.  -- Jayron 32 23:36, 17 May 2018 (UTC)
 * You have stumbled upon the property that lets Br− and I− act as effective nucleophilic catalysts; they really are both good nucleophiles and good leaving groups in SN2 reactions. It often improves yield in SN2 reactions to pass through an alkyl iodide on the way to the nucleophile you actually want, because an alkyl iodide is more easily substituted than an alkyl chloride (Clayden, Greeves, and Warren's Organic Chemistry discusses this point at the end of Chapter 15, which covers nucleophilic substitution at a saturated carbon atom.) Your reaction will not proceed to completion but will instead reach equilibrium. Other factors (such as the solvent) can drive such halogen exchange reactions to completion by giving things a push in one direction or the other; this is how the Finkelstein reaction works (although that is converting alkyl chlorides or bromides to alkyl iodides). Double sharp (talk) 15:28, 19 May 2018 (UTC)
 * P.S. If you can access it, 10.1021/ed074p836 discusses converting alkyl chlorides to alkyl bromides and vice versa. Double sharp (talk) 15:49, 19 May 2018 (UTC)

What Does Spironolactone have to do with Chronic heart failure that other diuretics doesn't have?
I saw the following question "Despite the administration of cardiotonic and thiazide diuretic a patient with chronic heart failure has persistent edemas and the risk of ascites arose. What medication should be administered to enhance the diuretic effect of the administered drugs? A. Manithol B. Furosemide C. Amiloride D. Clopamide E. Spironolactone". The correct answer according to books and sites is Spironolactone. My question is what does it have to do with chronic heart failure that the others don't have? All of them diuretics, some of them are weaker and some of them are more potent diuretics, but I don't understand why they expect from the answerer to answer Spironolactone exactly while the others can be apparently also correct answer. Isn't it? 93.126.116.89 (talk) 13:30, 17 May 2018 (UTC)
 * Even without any knowledge of the medications one can grok the reason if one is told the answer is Spironolactone. The others must be cardiotonics or thiazide diuretics, which aren't working. Spironolactone must work on another pathway. Abductive  (reasoning) 14:46, 17 May 2018 (UTC)
 * ) Spironolactone is Spironolactone is K+ sparing diuretic and the rest are cardiotonics as well... I know the pathways of each of this drugs and I don't find any reason for this to be the chosen answer logically. Sometimes there are MCQs that have mistakes that's why I'm asking. 93.126.116.89 (talk) 16:19, 17 May 2018 (UTC)
 * Spironolactone is a diuretic, but not a thiazide diuretic! It is a specific pharmacologic antagonist of aldosterone, acting at the collecting tubule, preventing the adaptive process of cell hypertrophy that occurs with long-term loop diuretic use. Circulating aldosterone concentrations are often increased in advanced congestive heart failure, which makes spironolactone yet more effective. Spironolactone opposes the kaliuretic effect of loop or thiazide-type diuretics. And the addition of this potassium-sparing diuretic offsets the potassium-wasting of loop and thiazide-type diuretics. And maintaining normal potassium levels can be important when digitalis (a cardiotonic) is being administered. But the answer is probably to test if you know the empiric evidence: that "spironolactone has been found to increase life expectancy and reduce hospitalization frequency when added to the conventional therapeutic regimen of patients with advanced congestive heart failure and systolic dysfunction.". - Nunh-huh 17:31, 17 May 2018 (UTC)
 * The question is really beating the point to death. For CHF patients, spironolactone is preferred. For patients with persistent edemas, spironolactone is preferred. For patients with ascites related to cirrhosis, spironolactone is preferred. So, the question gives the patient all three and asks which is preferred. 209.149.113.5 (talk) 19:51, 17 May 2018 (UTC)
 * No, that's not the case. Loop diuretics like furosemide are the mainstay of diuretic therapy in heart failure. Spironolactone is often added to loop diuretics, but it can't be simply stated that it is "preferred" in CHF. It is, as you say, the drug of choice in initial treatment of ascites due to cirrhosis. - Nunh-huh 22:16, 17 May 2018 (UTC)
 * I agree that furosemide is more effective diuretic in this context, but note that the question asks which would "enhance the diuretic effect of the administered drugs". Spironolactone, acting downstream (in the collecting duct) of the thiazides in the nephron would prevent the usual distal compensation for the more proximally-acting (distal convoluted tubule) thiazide. I also wonder whether the question might be a little dated, referring to spironolactone rather than newer more selective antagonists like eplerenone, and the thiazides rather than the more recent thiazide-like diuretics like indapamide. I agree with the K+/digoxin comments of Nunh-huh. Klbrain (talk) 22:30, 18 May 2018 (UTC)
 * Well, yes, but if the question is designed to elicit the fact that adding spironolactone to furosemide results in synergy rather than a merely additive effect, it would benefit from a rewrite. Your point about it being a dated question is well-taken, though my thoughts were that it was perhaps aimed at practitioners in areas where the latest medications are not available. - Nunh-huh 00:26, 19 May 2018 (UTC)

STO cells; what does STO stand for?
STO feeder cells are a cell line that are used to enable growth of other cells in culture. What does STO stand for? Does it even stand for anything? Everyone just seems to call them STO cells without explaining the letters. — Preceding unsigned comment added by 129.215.47.59 (talk) 15:31, 17 May 2018 (UTC)
 * Our STO page suggests it is " S IM T hioguanine/ O uabain-resistant mouse fibroblast cell line", which agrees with what 10.1095/biolreprod.103.017467 says about it and its relevance to your context. And clicking on the [characteristics] tab of the webpage you linked agrees that we're all talking about the same thing. DMacks (talk) 15:58, 17 May 2018 (UTC)

Joseph-Louis Lagrange
Hi,

I've just read the article about Lagrange, who is famous for the Lagrangian points. In the article, it says "before his death at Paris in 1813, in 128 rue du Faubourg Saint-Honoré". As a Frenchman, I'm very familiar with this street (this is our "Pennsylvania Avenue") and I wanted to find out more about his house in the Presidential Palace's street. Problem is, when you google "128 rue du Faubourg Saint-Honoré", there is nothing about Lagrange. Are you sure the information in the article is correct? Ericdec85 (talk) 15:40, 17 May 2018 (UTC)
 * He died at the end of the Napoleonic era - Rue du Faubourg Saint-Honoré wasn't really established as a major street yet, the whole area would be redeveloped under Baron Hausmann. So (like so much of Paris) the path of the street is ancient, but the buildings aren't. I would suspect that wherever he lived on this street, there's probably very little left of that specific address today. Andy Dingley (talk) 15:57, 17 May 2018 (UTC)
 * Organisation et reglemens de l'institut des sciences lettres et arts, Paris, Janvier 1807 (p. 267): "Liste des membres par ordre alphabetique - LAGRANGE (Joseph-Loius), rue du Faubourg Saint-Honoré, no.128". Alansplodge (talk) 17:49, 17 May 2018 (UTC)
 * However, Visiter quartier Saint-Honoré says: "No. 134 : Empire-style hotel built for the mathematician Joseph-Louis Lagrange and acquired in 1846 by the family of the banker Alexandre Aguado" (I couldn't stop Google from translating it). Alansplodge (talk) 18:15, 17 May 2018 (UTC)
 * This page also opts for no. 134 - perhaps he moved? Alansplodge (talk) 18:21, 17 May 2018 (UTC)
 * Or the number moved. Streets are sometimes renumbered.  Hard to say.  Regarding the location, the Élysée Palace was built in 1722, so it was certainly there the early 1800s, and in 1813, it was actually owned by Napoleon, according to our article.  However, that being noted, that doesn't mean that it was then, or is now, the only building on the street.  The analogy to Pennsylvania Avenue is not necessarily all that apt; yes, the White House has a Pennsylvania Avenue address, but that's a fairly long street, and it goes through many different neighborhoods, some of which are sketchy enough I don't think you'd want to be there after dark.  Anhyoo, back to the rue du Faubourg Saint-Honoré This is the approximate location of the address today; and it doesn't look like the kind of place where he wouldn't have lived; it's a good mile from the Élysée and there's lots of what looks like late 17th century buildings there.  It may even still be there today.  -- Jayron 32 18:58, 17 May 2018 (UTC)
 * This is no. 134 today. It is home to the Chambre de Commerce Italienne en France and a posh hairdresser. Alansplodge (talk) 19:44, 17 May 2018 (UTC)
 * Park Avenue in the 80s must've been something. Billionaires and Grand Theft Auto a few blocks apart. Sagittarian Milky Way (talk) 20:59, 17 May 2018 (UTC)

Can I say that every allergy is autoimmune disease?
If allergy is a wrong way of the body to interpret situations (such as considering nut as pathogens...) can all of allergies be under the umbrella of autoimmune system? And all of the autoimmune diseases are defect of the immune system in which the immune system doesn't interpret substances properly as it should. Isn't it? 93.126.116.89 (talk) 16:13, 17 May 2018 (UTC)
 * There's an immune system, but there's not an autoimmune system. Allergies are caused by the immune system (and can be further classified by the type of reaction that occurs—type I, type II, type III, or type IV hypersensitivities.) And yes, autoimmune disease also occurs via the action of the immune system. - Nunh-huh 17:50, 17 May 2018 (UTC)
 * No doubt the links above will give more specific information, but for a quick analogy, you can think of autoimmune diseases as being cases where your own tissues are directly hit by "friendly fire" from the immune system &mdash; cascades meant to destroy invaders instead destroy your own substance.
 * In an allergy, on the other hand, the immune system targets "invaders" that don't really need to be destroyed, and you suffer the negative repercussions of being in a "war zone".
 * For example, the IgE system is triggered by, say, a pollen grain, and it causes mast cells to degranulate, releasing histamine, which provokes local inflammation. If you had a parasitic worm trying to make its way into your flesh, that inflammation would presumably impede the worm somehow.  Pollen wasn't going to do that, so you didn't need the inflammation, and it causes you discomfort.  But the inflammation isn't the payload; it isn't meant to attack your tissues.
 * So no, I don't think they're the same thing. There's a reasonably clear distinction.  I don't know whether we have it laid out more expertly in our articles. --Trovatore (talk) 18:07, 17 May 2018 (UTC)


 * The answer is of course, no: In an allergy, the immune response is directed at an actual pathogen, it's just a benign pathogen, and any damage you suffer is due to the "secondary response" from the immune system. In an autoimmune disease, the auto is the key part: the immune system attacks your own cells as though they were pathogens.  Different thing entirely.  You can read about these differences at articles like allergy and autoimmune disease.  -- Jayron 32 18:48, 17 May 2018 (UTC)
 * So in other words, you could think of an autoimmune disease as being allergic to yourself. ;-)  2601:646:8A00:A0B3:7508:650A:FEA6:C618 (talk) 01:22, 18 May 2018 (UTC)

Yanny or Laurel
Is this the only known recording which some English speakers are sure is one word and others are sure is a different word or name? I wouldn’t include recordings which are simply too faint, distorted or drowned out by noise or music to hear clearly, like movie scenes which need captions to be understood. I would also exclude Mondegreens. Edison (talk) 21:10, 17 May 2018 (UTC)
 * There are other such auditory illusions, though they are often constructed in far more complex ways. The Tritone paradox is one.  If you want to learn more about the effect (the Yanny/Laurel effect or the Shepard tone effect at the center of the Tritone illusion) this is all part of the field of Psychoacoustics.  My suspicion is that the Yanny/Laurel thing is someone hacking into the brain's ability to filter sounds and focus on specific speech patterns, known as the Cocktail party effect.  This sort of auditory hacking is the the aural equivalent of optical illusions. -- Jayron 32 23:50, 17 May 2018 (UTC)
 * Maybe you could explain how anyone is hearing "laurel" in the recording in question. ←Baseball Bugs What's up, Doc? carrots→ 23:47, 17 May 2018 (UTC)
 * YOU apparently hear "Yanny". Other people hear "Laurel", in fact its pretty close to 50/50.  I can only hear "Laurel" in the original recording, and heat nothing of the Yanny.  It really is a weird effect, but its real.  Just a reminder that what goes on inside your brain (or my brain) is not universal to humanity.  All experience is unique.  -- Jayron 32 23:50, 17 May 2018 (UTC)
 * The clip they played on Fallon last night (assuming it's the right one) sounded more like if you started "Larry" with a "Y". I don't see how a trailing "ee" sound could be confused with a trailing "el" sound. ←Baseball Bugs What's up, Doc? carrots→ 00:33, 18 May 2018 (UTC)
 * The original sounds to me like Laurel, without question. The New York Times actually has a neat widget to help you hear it both ways, but honestly, it sounds like Laurel and Yarry to me, rather than Laurel and Yanney. https://www.nytimes.com/interactive/2018/05/16/upshot/audio-clip-yanny-laurel-debate.html Someguy1221 (talk) 02:26, 18 May 2018 (UTC)
 * This has some discussion and also a video with audio samples modified from the original clip which should enable most people to separately hear both Yanny and Laurel [//www.vox.com/2018/5/16/17358774/yanny-laurel-explained]. (The video is here on Twitter [//twitter.com/MBoffin/status/996562598815416321]) Nil Einne (talk) 02:17, 18 May 2018 (UTC)
 * I can hear Laurel quite easily. Run the sound through a graphic equaliser or similar sharp low-pass filter and strip the high frequencies.  Normally I hear "Yarry", but take the top end off and it does indeed turn into "Laurel". Andy Dingley (talk) 08:51, 18 May 2018 (UTC)


 * Here's a clip in which there is no ambiguity. The "yanny"/"laurel" version must have been technologically inferior. ←Baseball Bugs What's up, Doc? carrots→ 01:48, 18 May 2018 (UTC)


 * I can't believe that anyone hears this as "Yanny" ... and given that we know it is a recording of the word "Laurel", the other point of view seems hard to understand. But is the difference a matter of the human ear, or are some people listening on equipment that somehow confuses the sound? Wnt (talk) 02:19, 18 May 2018 (UTC)


 * People are able to hear both. Some people after trying hard enough (I even read of someone who tried hard enough and now can only here that), some people semi randomly. So while equipment likely plays a part, clearly the human auditory system (by which I'm including the brain) plays a big part. Nil Einne (talk) 02:27, 18 May 2018 (UTC)
 * The difference is you hear Laurel and someone else hears Yanny. You are not right and they are not wrong, and you are not wrong and they are not right.  Both sounds are present in equal measure, and how your own auditory processing system (brain and ears) deals with it is different than someone else.  That's just it.  It's nothing more than a lesson that your sensory perception is not universal. No one's is.-- Jayron 32 02:31, 18 May 2018 (UTC)


 * (ec, continuing...) But --- I just RTFA and was directed to this tool.  By pulling their slider over two units, I was able to hear "Yanny".  Pulling it back one unit (to +1 Yanny) made it sound like "Laurel" again ... but putting it to +1.1 Yanny and then slowly working it back, I could hear that version even at +0.9 Yanny units.  But after two minutes, trying again, I had to turn it up to +3 Yanny to hear that version, and tended to lose it around +1.2 or so.  The one thing that's clear from my perspective is that there is substantial hysteresis turning up and down ... but it is possible to hear the versions alternately, without adjusting the slider at all, at some point in the intermediate zone, like +1.6 is good.  Trying to switch interpretations, sometimes I "fail" and hear the previous one, but usually I make it sound like the version I'm trying to hear at that point.  I feel some kind of actual muscular tension/response in my ears doing this mental shift, especially trying to hear the "Y" start ... Wnt (talk) 02:32, 18 May 2018 (UTC)
 * And while I was posting that, and hit an edit conflict, with the sound repeating and getting distracted a moment, I somehow got stuck on "Yanny" so hard that I couldn't make it sound like "Laurel" without turning the slider back down to +0.8 or so. But a minute later I couldn't hear "Yanny" without a hard deliberate effort, same setting.  This is odd.  (Whether my hand is between the speaker and the ear seems to have some effect favoring Yanny, I think, but it isn't that strong)  Wnt (talk) 02:37, 18 May 2018 (UTC)


 * Sorry to be the bore, but this effect is neither special nor unique to this specific audio recording.
 * In the general case, you can take any two audible waveforms and combine them using a mixer. For any two waveforms, there exists a wet/dry mix such that any individual will preferentially identify the wet- or dry- as the perceptually dominant sound.  This is definitionally true: if we ask participants to classify a waveform as sounding like "A" or "B", there always exists some mix ratio for which there will be an exact 50-th percentile split in any focus group between those who hear "perceptual response A" and those who hear "perceptual response B."  In many cases, that split will occur when the mix ratio is also exactly 50% wet/dry; but it could occur at any other value.  With a very very trivial bit of standard off-the-shelf technology, and a large-enough focus group, we could construct such an "ambiguous" waveform from any two sounds or phonemes.
 * Here's a good article on mixing basics from the Music Department at University of Indiana.
 * And if you're in the mood for meatier treatment of acoustic processing theory, here's Physical Audio Signal Processing, one of my favorite technical books of all time. The chapter on voice synthesis gives a great overview on the history and technology of the vocoder and the art and practice of phoneme synthesis.
 * "One of the difficulties of formant synthesis methods is that formant parameter estimation is not always easy." This is techno-babble jargon that just means we all think we hear different things when we listen to the exact same waveform.
 * If you're really interested, another great text is Rabiner and Schafer, Theory and Applications of Digital Speech Processing....
 * If you start playing games audio filtering, humans may still recognize the sound, but will notice that it's not quite right. One of the easiest ways to disarm the audience, and to make them more receptive to silly audio processing tricks, is to lower the signal to noise ratio by intentionally injecting noise. In fact, a great many of the important historical voice synthesis algorithms begin by injecting pure white noise, and then progressively filtering that noise until it sounds like human speech!  The practicality of this is that if you use a modern operating system, you can make the computer talk; and you can change the voice.  Ever wonder why there's almost universally an option to "whisper", in almost every configurable text-to-speech synthesizer?  That synthesized breathy voice is full-mix white-noise injected into every synthetic phoneme!
 * Nimur (talk) 03:32, 18 May 2018 (UTC)
 * If this is easy to do by mixing audio or filtering white noise or employing psychoacoustics or whatever, then per the original question where is another audio sample which about half the people hear as one word and the rest hear differently, and can’t Imagine how the others can hear what they hear. When this phenomenon first sprang up I seriously wondered if it was a gag and my signifcant other was just pretending to hear Yarry when I heard Laurel. I doubt it is a function of people with more sensitivity to high frequencies hearing Yanni since an 80 year old and someone with significant heating loss heard Yanni or Yarry or Yammy but my child heard Laurel. This makes me wonder if it is genetic.  Edison (talk) 04:11, 18 May 2018 (UTC)
 * It is rather common. It is more common with digital media. The recording in question is a recording of the word "laurel." That is well documented, including the fact that it was made by an opera singer because the company that hired the speakers preferred opera singers. If you were to listen to the person actually speaking, you would hear laurel. After being digitized and compressed, some people year "yanny." If you go to music recordings, which are being digitized and compressed, you find the same thing. There are some words or phrases that, even without compression, are difficult to understand (e.g. "excuse me while I kiss this guy"). There are also many single words that sound like other words. I don't want to start a long exhaustive list, but everyone knows some lyric where a word just doesn't sound like it should. For me, personally, mp3 versions of Stairway to Heaven really mess up the end. "How everything turns to gold" really sounds like "How everything turns to stone." 209.149.113.5 (talk) 12:09, 18 May 2018 (UTC)
 * Here's a multimedia clip of some historical synthesis: Bell Labs' The Voder (1939), including some comparative phoneme modes. Suffice to say, not every listener thought it sounded like it's speaking English.  And when it speaks French... well, I'm sure somebody thinks it's intelligible.  (Thus began the storied history of annoying robots making barely-intelligible telephone calls, and the many generations of naive scientists who think they invented it, even though people have not wanted this technology for over a century).
 * If you want more examples, have a listen to the radio traffic at a busy airport on a website like LiveATC. Especially if there's radio static, the words can be pretty hard to distinguish.  By the time the VHF voice radio signal gets received, processed, and digitally compressed for internet transmission, a lot of phonemes sound tricky.  "Zero" can sound like "four"; "departure" can sound like "SouthWest".  If you spend enough time listening to, or using, voice radio, you'll develop some skills to help reduce collision - but you'll also find times when you need things to get repeated!
 * If you're looking for more sample::: audio files, here's a whole website dedicated to ultra-low-bit-rate voice compression: Rowetel: open telephony software. Listen to the 700 bit-per-second English speech recording: .  The limit of ultra-high compression ratio is the point where each phoneme is "one unit of JND" above un-intelligble!
 * Nimur (talk) 15:00, 18 May 2018 (UTC)
 * Audio files which are simply distorted or degraded and unintelligible but do not present seemingly clear and unconfusable but different perceptions are not what I am looking for. Where is an audio file which some hear as “X” and others hear as an unmistakable “Y” where X and Y are different words or names. I have only seen assertions that it is easy and common, without any examples. In visual science there are many common examples such as the image which is a beautiful young woman or an ugly old woman, or the duck/rabbit, or the Necker cube. The Shepard ascending tone does not sound like a descending tone to half the listeners. Mondegreens do not count. Edison (talk) 02:11, 19 May 2018 (UTC)
 * There are some great answers here, but there are some specifics not touched on. Why "Laurel" and "Yanny" but not intermediate forms?  I mean, "L" vs. "Y" would seem to be a matter of whether the tongue touches the palate, and so is "R" vs. "N"... but in the opposite direction!  So why no "Yannel", "Laurie", "Yaurel", "Lanny"...?  How can it be reliable that I, hitting this for the first time, pick out these two and only two combinations, the same as everyone else, when I shift my mind listening to the same ambiguized version of the word repeated over and over? Wnt (talk) 08:33, 19 May 2018 (UTC)
 * Well if you read the sources, some people do say to hear something slightly different. In fact even in this thread we have someone saying Yarry. I suspect these have at least in part to do with what accents and other stuff people understand and expect. I mean for most people WTF is a Yannel, Yaurel or Lanny? More to the point though, I'm not sure why you're still doing weird things like analysing this from tongue movements. There is extensive discussion of the different frequencies involved and how they are perceived in the sources. These are what matter, not tongue movements. At least partially understanding that would likely be a key part of answering this question, since most people seem to be hearing either the low frequencies or the high frequencies rather than tongues in different positions. How these frequencies are normally produced in human speech seems to me to be largely an aside except getting back to the earlier point on what people understand and expect. Nil Einne (talk) 01:40, 20 May 2018 (UTC)