4 ms·

> Every professional is able to distinguish a Stradivarius from a modern instrument. The available evidence does not support your assertion, in fact it refutes

> Every professional is able to distinguish a Stradivarius from a modern instrument. The available evidence does not support your assertion, in fact it refutes it. I mean, you can assert however much you like, but anyone who suscribes to the scientific method would disagree based on the studies. (EDIT: That's not say that I don't "get" why people would want a Stradivarius -- its historicity, its rarity, etc. etc. Sound, however, cannot be the distinguishing feature.)

owenversteeg 10y ago
[edit] hm, +2 to -4 in quick succession on a dead post. That's interesting. TL;DR The published research is absolute worthless garbage, with trash statistics, re-used participants, and downright inflammatory and confusing quotes such as the following: "Some musicians espouse decidedly nonscientific views, such as the existence of spirit guides". Say what you will about Stradivarius violins - I myself prefer Il Cannone to any Strad - but please do not try to prop up an argument with these worthless studies. Your note of "Anyone who suscribes [sic] to the scientific method would disagree based on the studies" irritates me. "subscribing to the scientific method" is related to accepting published results in the same way that iPhone sales are related to the number of people who died falling down stairs [1]. Even more irritating, if you actually took two minutes to look for real, published studies you'd notice that the literature is surprisingly lacking for such a blogged-about topic. The entirety of published, close-to-legitimate research consists of two studies. Upon a closer look, one may notice that both studies were published in PNAS through the "Contributed" process, allowing them to bypass most parts of peer review - and critically, select their reviewers. After scanning the headline, you may notice that four of the five authors on the original paper are present on the second. Simply Googling the name of the first paper indicates strong concerns that have been voiced about the experimental design of the first paper, which made even the newspapers that published the second paper call the first paper "controversial". Then, when you look at the second paper, you'll find a number of concerning things. I'm not sure how many papers you write, but even if the number is small I'm sure you'll agree that "The Team" is not something you'll find often in published literature. I assume one of their cherry-picked reviewers realized this, and thus they added the sentence "Although it is unusual to describe the team..." right after starting their "The Team" section, which made me chuckle. Once you start actually reading the beef of the paper, things start to look much more concerning. 40% of the violinists voiced deep concerns with the study. What really makes me laugh was the supposed selection process. They claim to have selected a good sample of violinists (and make the, er, interesting claim that their sample is equivalent - yes, equivalent - to the entire population.) Then, only in the Supporting Information - accessible only by a link at the end of the paper in small text, by the way - they disclose that some of the violinists that were used in this particular study were exactly the same violinists used in the previous study. This becomes even more concerning when you realize the sample size is 10. A sample size of 10 is not bad by itself, and can often be justified, but in this case they reused study participants who had played the same violins before, and then in the same breath boasted about the quality of their study participants. I'm not sure how this paper got published a SECOND time. Oh wait, I forgot. PNAS Contributed Papers. Letting Bad Science Bypass Peer Review Since 1914!™ For a fun time (warning: your definition of fun may be different than mine) you should read the absolutely disgusting commentary article [2]. It's basically an essay by the author on "Why Science Trumps Silly Artists!". It includes things such as hilariously saying that irrational minds will see triangles in Kanizsa's triangles when (and I quote) "careful inspection" reveals none actually exist! Wow! Scientists™ realized Kanizsa's triangles don't actually have triangles?! I for one am glad that we have Scientists to tell us that! There's no way our feeble artist brains could figure that out ourselves! The author then continues to let free-form bullshit dribble onto the page. Pop quiz: Are these quotes from Deepak Chopra, or a scientist with a PhD published in PNAS? - "Expectations retune neural circuits" - "Our processing can change perception" If you answered Deepak Chopra, you're wrong! These are actually quotes from a PhD professor published in PNAS. He goes on to say some downright inflammatory and ridiculous things: - "The findings should put an end to the outrageously high prices charged for musical instruments" What? I really hope this guy doesn't actually believe what he writes. - "Because artists rely so heavily on their own experience, studies like this have historically fallen on deaf ears—“I know what I know because my senses tell me so” may be the refrain of those who are skeptical of scientific methods." Excuse me? The singular previous study was discredited by scientists first, then musicians second, because of sloppy science. - "Some musicians espouse decidedly nonscientific views, such as the existence of spirit guides" What? I'd like to hear this man try to justify the inclusion of this sentence. Exactly what was he trying to accomplish? I'm shocked and appalled at the current published "research", if it can even be called that. Say what you will about Stradivarius violins - I myself prefer others - but please do not defend the "research". [0] https://xkcd.com/1425/ https://xkcd.com/1425/ [1] http://tylervigen.com/view_correlation?id=28669 http://tylervigen.com/view_correlation?id=28669 [2] http://www.pnas.org/content/111/20/7168.full http://www.pnas.org/content/111/20/7168.full [3] for more notes on violin stuff in general, see http://pastebin.com/GCuK322j http://pastebin.com/GCuK322j
- lomnakkus 10y ago
  I can only guess that the downvotes you're getting are based mostly on the "tone" of your reply -- which is pretty hostile, but whatever. Ok, so here's the deal: If it's really easy to get these things published, then I suggest (if you care enough, you clearly care a great deal) that you try to get a paper puplished on your double-blinded randomized trial (etc.). Honestly, it sounds like you might have a beef with the scientific process[1]. Btw, what's going on with your (3) link? Was that your edit buffer or something? [1] No, it's not flawless. There's publication bias, etc., but so far it's the best anyone's come up with.
  - owenversteeg 10y ago
    My apologies if it seemed hostile. I get worked up about bad science, and PNAS's process in particular is very bad-sciency, allowing anyone famous enough to bypass the peer review process. Nothing about you particularly, or your comment :) And hah, no beef with the scientific process here. I'm definitely very pro-science, and pro-scientific process. But that also makes me very anti-bad-science, and that paper is definitely a jumble of bad science that somehow scraped by the peer review process because of nepotism in PNAS. Yep, [3]'s my 3am brain dump. Mostly unrelated but dumped in that comment cause it's the only one I can edit. And now you've got me thinking about how feasible it would be. If I actually go through with this, I'm blaming you for it :)
    - lomnakkus 10y ago
      Heh, happy to take the blame/accolades for the inspiration! :) I definitely think there's room for improvement (no particular experience with PNAS) in general -- peer review is very flawed, and there are huge problems with publication bias, the area being so specialized that the reviewers (indirectly) know whose article they're reviewing, etc. Actually, I think one of the most interesting things (science-wise) to come out in recent years is the meta-science of science-about-science, fuzzy as that may be. I'm guessing you already know of him, but Ben Goldacre has been a great source (for us outsiders) for information on this.

owenversteeg 10y ago

[edit] hm, +2 to -4 in quick succession on a dead post. That's interesting. TL;DR The published research is absolute worthless garbage, with trash statistics, re-used participants, and downright inflammatory and confusing quotes such as the following: "Some musicians espouse decidedly nonscientific views, such as the existence of spirit guides". Say what you will about Stradivarius violins - I myself prefer Il Cannone to any Strad - but please do not try to prop up an argument with these worthless studies. Your note of "Anyone who suscribes [sic] to the scientific method would disagree based on the studies" irritates me. "subscribing to the scientific method" is related to accepting published results in the same way that iPhone sales are related to the number of people who died falling down stairs [1]. Even more irritating, if you actually took two minutes to look for real, published studies you'd notice that the literature is surprisingly lacking for such a blogged-about topic. The entirety of published, close-to-legitimate research consists of two studies. Upon a closer look, one may notice that both studies were published in PNAS through the "Contributed" process, allowing them to bypass most parts of peer review - and critically, select their reviewers. After scanning the headline, you may notice that four of the five authors on the original paper are present on the second. Simply Googling the name of the first paper indicates strong concerns that have been voiced about the experimental design of the first paper, which made even the newspapers that published the second paper call the first paper "controversial". Then, when you look at the second paper, you'll find a number of concerning things. I'm not sure how many papers you write, but even if the number is small I'm sure you'll agree that "The Team" is not something you'll find often in published literature. I assume one of their cherry-picked reviewers realized this, and thus they added the sentence "Although it is unusual to describe the team..." right after starting their "The Team" section, which made me chuckle. Once you start actually reading the beef of the paper, things start to look much more concerning. 40% of the violinists voiced deep concerns with the study. What really makes me laugh was the supposed selection process. They claim to have selected a good sample of violinists (and make the, er, interesting claim that their sample is equivalent - yes, equivalent - to the entire population.) Then, only in the Supporting Information - accessible only by a link at the end of the paper in small text, by the way - they disclose that some of the violinists that were used in this particular study were exactly the same violinists used in the previous study. This becomes even more concerning when you realize the sample size is 10. A sample size of 10 is not bad by itself, and can often be justified, but in this case they reused study participants who had played the same violins before, and then in the same breath boasted about the quality of their study participants. I'm not sure how this paper got published a SECOND time. Oh wait, I forgot. PNAS Contributed Papers. Letting Bad Science Bypass Peer Review Since 1914!™ For a fun time (warning: your definition of fun may be different than mine) you should read the absolutely disgusting commentary article [2]. It's basically an essay by the author on "Why Science Trumps Silly Artists!". It includes things such as hilariously saying that irrational minds will see triangles in Kanizsa's triangles when (and I quote) "careful inspection" reveals none actually exist! Wow! Scientists™ realized Kanizsa's triangles don't actually have triangles?! I for one am glad that we have Scientists to tell us that! There's no way our feeble artist brains could figure that out ourselves! The author then continues to let free-form bullshit dribble onto the page. Pop quiz: Are these quotes from Deepak Chopra, or a scientist with a PhD published in PNAS? - "Expectations retune neural circuits" - "Our processing can change perception" If you answered Deepak Chopra, you're wrong! These are actually quotes from a PhD professor published in PNAS. He goes on to say some downright inflammatory and ridiculous things: - "The findings should put an end to the outrageously high prices charged for musical instruments" What? I really hope this guy doesn't actually believe what he writes. - "Because artists rely so heavily on their own experience, studies like this have historically fallen on deaf ears—“I know what I know because my senses tell me so” may be the refrain of those who are skeptical of scientific methods." Excuse me? The singular previous study was discredited by scientists first, then musicians second, because of sloppy science. - "Some musicians espouse decidedly nonscientific views, such as the existence of spirit guides" What? I'd like to hear this man try to justify the inclusion of this sentence. Exactly what was he trying to accomplish? I'm shocked and appalled at the current published "research", if it can even be called that. Say what you will about Stradivarius violins - I myself prefer others - but please do not defend the "research". [0] https://xkcd.com/1425/ https://xkcd.com/1425/ [1] http://tylervigen.com/view_correlation?id=28669 http://tylervigen.com/view_correlation?id=28669 [2] http://www.pnas.org/content/111/20/7168.full http://www.pnas.org/content/111/20/7168.full [3] for more notes on violin stuff in general, see http://pastebin.com/GCuK322j http://pastebin.com/GCuK322j

lomnakkus 10y ago
I can only guess that the downvotes you're getting are based mostly on the "tone" of your reply -- which is pretty hostile, but whatever. Ok, so here's the deal: If it's really easy to get these things published, then I suggest (if you care enough, you clearly care a great deal) that you try to get a paper puplished on your double-blinded randomized trial (etc.). Honestly, it sounds like you might have a beef with the scientific process[1]. Btw, what's going on with your (3) link? Was that your edit buffer or something? [1] No, it's not flawless. There's publication bias, etc., but so far it's the best anyone's come up with.
- owenversteeg 10y ago
  My apologies if it seemed hostile. I get worked up about bad science, and PNAS's process in particular is very bad-sciency, allowing anyone famous enough to bypass the peer review process. Nothing about you particularly, or your comment :) And hah, no beef with the scientific process here. I'm definitely very pro-science, and pro-scientific process. But that also makes me very anti-bad-science, and that paper is definitely a jumble of bad science that somehow scraped by the peer review process because of nepotism in PNAS. Yep, [3]'s my 3am brain dump. Mostly unrelated but dumped in that comment cause it's the only one I can edit. And now you've got me thinking about how feasible it would be. If I actually go through with this, I'm blaming you for it :)
  - lomnakkus 10y ago
    Heh, happy to take the blame/accolades for the inspiration! :) I definitely think there's room for improvement (no particular experience with PNAS) in general -- peer review is very flawed, and there are huge problems with publication bias, the area being so specialized that the reviewers (indirectly) know whose article they're reviewing, etc. Actually, I think one of the most interesting things (science-wise) to come out in recent years is the meta-science of science-about-science, fuzzy as that may be. I'm guessing you already know of him, but Ben Goldacre has been a great source (for us outsiders) for information on this.