|This is the talk page for discussing improvements to the Absolute pitch article.|
|WikiProject Music theory||(Rated B-class, Mid-importance)|
|Wikipedia Version 1.0 Editorial Team / v0.7|
- 1 Upgrade rating
- 2 Nature vs Nurture
- 3 Contextual pitch memory
- 4 History of study and terminologies
- 5 Recently arisen evidence or new locus?
- 6 Benefits and detriments
- 7 So how perfect is "perfect?"
- 8 Mechanism?
- 9 Accomplished/Important musicians not having it.
- 10 Suggested move
- 11 Pseudo AP
- 12 Linguistics - East Asia
This is the best material I have read on absolute pitch. It is certainly an example of good Wiki scholarship. We might want to consider promoting it. Piano non troppo (talk) 17:07, 29 April 2010 (UTC)
Nature vs Nurture
- Yeah, I havn't seen the previous discussion. Do you have evidence that the two studies I cited were "fatally flawed", or do you just assume so? Gregcaletta (talk) 04:52, 14 May 2010 (UTC)
- The "fatal" flaw in the experimental design of both Nering's thesis and Rush's dissertation is that they only had two groups: training and no-training. Every study back to 1899 shows that anyone can gain some note-naming ability irrespective of the "method" employed. Therefore it would be necessary, in studying the effectiveness of a particular training method, to have a third group trying to memorize notes using a different system, or a fourth group trying to learn but with no "system" at all. By statistically comparing these groups to the group using the method under consideration, it could be seen whether any success in note-naming could legitimately be attributed to the method or if it was just because people were making an effort. Without this comparison, and in the face of all prior evidence, it cannot be concluded from either Nering or Rush's work that the training method caused the results.
- Secondly, but no less fatally, the data blatantly fail to support the conclusion that anyone "learned perfect pitch." Rush's study is available on-line... if you look at that study, and look at the actual data results on page 154, you see that the experimental subjects were, in post-test, able to correctly name an average of 25 out of 120 notes (20%). While this is an improvement from the pre-test, and it's better than the control group, it doesn't quite compare to the 120 out of 120 typical of absolute pitch ability.
- Rush separates out the best performers (page 194) because it's interesting, but it's not statistically valid. Imagine testing a new drug on 26 patients; of these, 4 show improvement and 1 seems to get well. If you ignored the 21 who didn't get better, and looked at the results only for these healthy five, it might look like the drug cured them-- but there are 21 failures whose data say otherwise. Even then, I'd draw your attention to Rush's actual numbers on page 202. If you take out the best performer, "Subject T," the average of the remaining 4 patients drops to 47 out of 120 correct responses-- and then if you look at the self-reports on page 400 you'll see that "Subject T" attributes his success to having abandoned the Burge training midway through (in favor of a melody strategy, like Pitch Paths, and this fact is buried in Rush's writing somewhere but I don't feel like searching for it right now).
- Nering's data are no different. Her experimental participants achieved 30% accuracy in naming notes after a year of training. Any of Nering's conclusions about "learning perfect pitch" suffer from the same errors (lack of a second experimental group and post-hoc sampling bias).
- So why not include these results in the article, and the reader can decide what to make of them? You are right that these results don't prove that everyone who attempts the Burge program will gain perfect pitch or that the Burge program is better than other programs. However, they do show that it is possible to develop the ability to some extent in most people and to a large extent in some people, and isn't this worthy of inclusion in the article? Gregcaletta (talk) 02:44, 15 May 2010 (UTC)
- The reason not to include these two studies is right there in your question: "these results". The only valid statistical result in either of these studies is that "people who practiced naming notes got better at it," and there are published studies that already show that result-- Cuddy and Meyer immediately spring to mind. I agree that this point is worthy of inclusion... and that's odd... this article used to say this, somewhere, and now it doesn't. But these two unpublished studies are not the sources to use. I don't know where the old discussion got archived to (or where any old discussion gets archived to) but it would probably save time and repetition if you dug it up and checked it out. aruffo (talk) 04:24, 15 May 2010 (UTC)
OK. If you could help me find those two particular studies that would be appreciated. However, my main problem is not the omission of this particular fact, but rather the fairly unjustifiable statement which follows in the paragraph, "no adult has ever been documented to have acquired the ability", a slightly toned down version of what it actually says in the cited article, "no adult has ever acquired the ability". I mean, what proof could they possibly have for that? The answer is they can't because it is impossible to prove a negative. So wouldn't you say this statement is "fatally flawed" also? Gregcaletta (talk) 11:40, 15 May 2010 (UTC)
- You'll be surprised.. the statement is actually true and justifiable. The published literature on absolute pitch is an astonishingly small body of work. The statement about no adults learning perfect pitch is easily supported by reading every document ever published about absolute pitch (which I have done, and which only takes a few months to do). The writers who say that no adult has ever acquired the ability are writing to a scientific audience who will understand that the "documented" is implied-- that no matter where anyone looks in the scientific literature, no evidence will be found to contradict the statement. You're right that this doesn't necessarily mean that no adult has ever learned perfect pitch, but if some adult has, it's never been documented anywhere. (Cuddy's article and Meyer's) aruffo (talk) 02:05, 16 May 2010 (UTC) (belated signing)
- Interesting. I don't have absolute pitch and I have tried the Burge program. I have to say I have to say that I think it would be possible to develop the ability using the method; I just think there are almost no adults would have the patience for it. If you do the method properly, you are supposed to do at least 10 minutes practice a day, along with your normal music practise, and you are supposed to spend as many days as you need on one lesson until you can identify 20 notes correctly in a row. The lessons start very simple, using relative pitch, and work there way up in difficulty until you are tested on all 12 notes including chords played rapidly. In other words, it would be impossible to get to the end of the course without developing the ability; it might just take years to do so. I worked on the exercises for a few months, but then travelled overseas and didn't have access to an instrument, and haven't been bothered to start over gain. I had already started to become aware, though, that each note has a particular quality to it, it's just very subtle. I just find it strange that the scientific community seems to assume that it is impossible to acquire the ability. I mean, even if you worked on it for hours each day for years in a row, I find it hard to believe that you wouldn't be able to develop it up to the level of the ability in "naturally occurring" cases. But even if you didn't acquire the ability, you still wouldn't have proved that it was impossible to do so, and yet that seems to be what the scientific community (and this article) assumes. Gregcaletta (talk) 09:13, 16 May 2010 (UTC)
- It may seem that people are assuming that, but it's not so. The sole scientific fact is that no evidence exists to show that any adult, anywhere, at any time, by any method, has learned absolute pitch ability. While this is obviously not an encouraging fact, it is not conclusive. A person who honestly believes that it is impossible to learn absolute pitch is ignorant of the facts. They deserve the reply given by Westley in The Princess Bride: "You're just saying that because no one ever has." No credible scientist would say outright that absolute pitch cannot be learned.
- Whether or not a method seems plausible in its approach is immaterial, as it is pure opinion. You can say that you think a method will probably work; another person can say they think the method probably won't work. The only fact of the matter is that none of them has ever delivered. Every scientist who has ever tested any adult learning perfect pitch-- including Rush and Nering-- has shown that note-naming ability can always be improved, but not to the extent of acquiring "unqualified" absolute pitch ability.
- As I recall, prolobe.com posts statistics from its members-- while this is not controlled scientific data-gathering, it might give you an idea of what people are up to. Prolobe users are fairly persistent. aruffo (talk) 21:32, 16 May 2010 (UTC)
- Yeah, fair enough. I just think it is unfortunate that "no adult has ever been documented to have acquired the ability" seems to imply that this is not simply due to the fact that there is a lack of documentation in general, although there is a lack. That prolobe.com website seems to have lots of members who have acquired the ability, some to a very high consistency, and yet we can't include it because no scientist has ever tested one of them beforehand, and then again afterwards, and then had the results published. Or hypothetically, hundreds of people could have acquired the ability using the Burge method, and it still would not be "documented". Gregcaletta (talk) 02:09, 17 May 2010 (UTC)
- Well said. That's the problem in one. ...Now that I think about it, I tried an experiment when I was at Indiana University in 2007, and I've got the data somewhere. It was the first experiment I conducted, so the design was very poorly controlled, but I tried to train people to learn absolute pitch using a melody system, and found essentially what I expected.. everyone learned to name notes, but the musicians did far better at it, and the non-musicians found the whole process too tedious to keep going. I wonder where I saved that data..? aruffo (talk) 04:23, 17 May 2010 (UTC)
Argh.. I found the data, and I'm disappointed that I didn't know enough about scientific procedure, when I was running it, to realize that I would definitely want two control groups-- one for musicians, one for nonmusicians. The control group of nonmusicians (who just practiced naming notes) improved slightly, but this was not significant. The experimental group of nonmusicians (who used melodies) also improved, and this improvement was by itself significant-- but not significantly different from the control group. The musicians improved more than the nonmusicians or the control group... but the difference between their improvement and the nonmusicians' is so wide that it invalidates any comparison between the musicians and the one control group (who were all nonmusicians). Drat! aruffo (talk) 16:40, 17 May 2010 (UTC)
- Sounds a bit complex for me. I'm not exactly sure what it shows but interesting results anyway. Gregcaletta (talk) 04:05, 19 May 2010 (UTC)
- I'm surprised though that some of the most high profile studies seem flawed even to me (I haven't finished my degree yet). For example the California study, they say "we also observed that absolute pitch aggregates in families, indicating a role for genetic components in its development". It doesn't necessarily indicate that at all, surely? I mean, an equally valid (and simpler) explanation for the fact that it aggregates in families is the parents are musical, and thus initiate the children into music at an earlier age, greatly increase the chance that they will develop perfect pitch. I also kind of find it kind of funny that they say "we learned that absolute pitch (perfect pitch) ability is a discrete perceptual trait, not simply the one end of a continuous "normal" distribution of pitch ability". This is a valid finding, I guess, but I would have thought it was obvious by the definition of absolute pitch. I mean, either you can tell the difference between the tastes whisky and rum, or you can't; it's not going to fall on a "continuous" scale or a "normal distribution". And I get the impression they think that this fact of discreteness too implies a some kind of binary deterministic gene. Gregcaletta (talk) 04:20, 19 May 2010 (UTC)
- Yes.. for a long time I've thought "genetic" absolute pitch research was silly for exactly that reason. As recently as 2002 (Brown et al, "Early Music Training and Absolute Pitch", Music Perception 19, 595) people were still making the suppositional argument "well you don't need musical training to have AP, therefore AP must be genetic." Recently, though, somebody did something worth mentioning.. now that I think about it, its findings should probably be written into this article (aren't they already? ...no, they don't seem to be). Gitschier and some others located different families whose members had AP and, comparing their DNA, found a correlation between them in a particular gene. I don't understand exactly how their research works, but an abstract can be found here. On the one hand, I would be willing to believe that there is a genetic component, especially if that genetic component promotes the hyper-development of the left planum temporale... on the other hand, the same gene they identify as associated with AP ability is also associated with susceptibility to colorectal cancer, so go figure. aruffo (talk) 04:45, 19 May 2010 (UTC)
- Yes, I think it is highly likely that the variation of genes would create a variation amongst brains and that some brains would therefore be slightly susceptible to the acquisition of absolute pitch. However, I would expect overall significance of this factor to be very slight, perhaps even negligible. The fact that the California study assumes a genetic link, and then after having assumed its significance, sets out to verify the existence of the link, which makes the study hugely susceptible to confirmation bias. Also, I find that in general with establishing genetic links there is a huge tendency towards assuming that correlation necessitates causation. For example, in an international study there would certainly be a correlation between asian genes and absolute pitch but this is more simply explained by the fact that they speak tonal languages; it does not necessarily imply any causal link. The same could happen with a US study. For example, if wealthy people are more likely to acquire the ability due to being more able to afford instruments and tuition, then you would find a correlation between the ability and the genes of ethnic groups which happen to have higher incomes on average. Again, it does not imply a genetic link.
- However, the most frustrating thing is that even if there is a significant genetic link -- it certainly at least remotely possible -- would it not be far more useful and interesting to focus the immense resources of the University of California or the effectiveness of certain methods or environmental factors, particularly among young children? Anyway, I think I have seriously violated WP:notaforum by now, but please let me know if any more evidence on the "environmental" side comes to your attention! Gregcaletta (talk) 05:27, 19 May 2010 (UTC)
Contextual pitch memory
How do studies of contextual pitch memory fit into this article? I mean studies that show that people consistently sing songs from memory in the same key that they learned them. Daniel Levitan discusses his experiment where subjects were asked to sing their favorite songs from memory; they consistently sang these songs at or very near the original pitch of the song (This is Your Brain on Music, p. 149). Blacking and other researchers have seen this phenomenon in other cultures as well.
I also wonder about instrument-specific pitch memory. My brother, a professional pianist, can identify any note when played on a piano. But he can't reliably identify a pitch when played on another instrument (violin, or wind instrument). Has this ever been studied?
- Yeah, this study (http://perfectpitch.ucsf.edu/study/). I think it still counts as having perfect pitch (if he really can identify the notes consistently on the piano). I would have thought he could learn to identify the notes on other instruments fairly easily if he practices it. Gregcaletta (talk) 04:01, 19 May 2010 (UTC)
- The question about "musical context" is a different one. This article shows there have been studies showing that most people have some kind of absolute reference for remember the keys that their favourite songs are played in, it just doesn't count as Perfect pitch unless they can consistently identify notes when played at random. Gregcaletta (talk) 04:03, 19 May 2010 (UTC)
History of study and terminologies
Hello. After saving this comment, I'll save a change to the article, comprising a new sub-section on the history of study and terminologies. On the whole, I like this article. And I especially like the liberal utilisation of references. There is almost always far more information on a given subject than can possibly be cited, so a range of material to launch into further exploration if I choose to or need to, is one of the criteria by which I judge an article or book. On looking at this article, I noted a citation tag next to the statement that recording (or documenting) of absolute pitch became more common for musicians of the 19th century. Two things occurred to me regarding the sentence and the tag. Firstly, it occurred to me that the tag may be there because the editor interpreted the sentence to mean that audio-recording of 19thC musicians became more common ('recording' having become narrower in contemporary popular meaning than the broader use which an academics or older generations might use the term). However the correct interpretation is of course 'documented', and I thought to change the term so there is no misunderstanding. But this would not change the underlying challenge, that of providing some sort of citation to the effect that from a certain era in history, documentation of absolute pitch became more common. And with this challenge, comes related questions, such as when did the term absolute pitch come into use? Was it circa 19th century? Was it recognised but known by different terminology prior to that? Etc.
This leads to the new sub-section which I have created. Research, ideas and knowlege always have a history, and knowing something of that history is often very enlightening, including opening up awareness of material one never previously realised existed, and which are germane to the subject one is studying. So I created the section and added a few citations. It will be possible to expand it further, but I don't have the time. Others may want to, and I would advocate that. But even should it remain undeveloped, at least there is something there. Regards Wotnow (talk) 00:39, 25 August 2010 (UTC)
Recently arisen evidence or new locus?
See this diff. A glance at the abstract of the reference does not give any hint that this is a newly arisen locus. I suspect that the evidence is the newly arisen item. __ Just plain Bill (talk) 00:24, 6 October 2010 (UTC)
Benefits and detriments
The "possible problems" section was getting confused with the "musical talent" section... I disambiguated the two in a way which would leave in the new Deutsch reference. aruffo (talk) 02:58, 6 October 2010 (UTC)
So how perfect is "perfect?"
With all the science lingo here, I see mostly music-talk, and a lack of sound-talk. (perception V. physics) Not a problem in itself, but... I believe that with a tuning fork or electronic tuner, any dummy can tune say, a guitar to within 1 Hz or better? Can a person with absolute pitch do that? (That question's why I read this article.) I see one place that could imply that 1/4 tone (ho-hum, often over 100 Hz?) is considered "perfect." If the answer is buried somewhere in the article, I think it should (also?) be in the opening definition paragraph. While "tone" and "pitch" to me imply a specific frequency, this is seemingly not part of their definitions, (following the hypertext). This needs to be spelled out or expanded. —Preceding unsigned comment added by 126.96.36.199 (talk) 14:12, 12 January 2011 (UTC)
YES. I would augment the above with the following: The article asserts that (without needlessly getting into the difference between pitch and frequency) possessing absolute pitch means that someone can identify a pitch from the current western twelve-tone scale, and that, in so doing, recognizes that say, 440 hertz is A4, and vice-versa. There are two obvious idiocies with this definition. First, why would absolute pitch have anything to do with a western twelve-tone scale? And second, in the real world (where most of us dwell), we know that as the difference between two things becomes smaller, the mechanism to recognize the difference must become more precise (and in most cases it takes longer to decide as well). Not to mention that a second is a completely arbitrary unit of time, therefore Hertz is a completely arbitrary unit of frequency.
So it is pretty obvious that what is really going on (my own wild assertion based on my own experience with music and physics) is humans (and probably other animals) vary in ability to recognize absolute pitch. Humans who can recognize absolute pitch within say, 20 cents (the musical kind--see the wiki page), and have the training to label pitches according to the western twelve-tone scale based on A4 = 440 hertz, meet most or all of the colloquial definitions in this article. But there is no clear line dividing humans into two groups as the article suggests, because there is no exact number of cents or labeling ability that establishes such a line. Mdlayt (talk) 01:35, 9 August 2012 (UTC)
Is it known how perfect pitch works? Have any experiments been done on how to recalibrate? For instance, a flute can be moved out of tune by changing temperature, or gas-mix. What environmental changes would make someone with perfect pitch go out of tune?
Also, how do different musicians perform together when their internal tunings are different? (for example, A440 and A442). —Preceding unsigned comment added by 188.8.131.52 (talk) 06:20, 10 February 2011 (UTC)
Accomplished/Important musicians not having it.
The article lists some examples of musicians that are known or at least it is documented that they probably had absolute pitch recognition. It would be good if some examples can be given of the opposite phenomenon. Are there well-known accomplished musicians that do not have/had the ability? This not only fills a possible curiosity of some people (me for example) but also shows to what extend it is not known (as it is written in the article) the correlation between absolute pitch and musical accomplishment. — Preceding unsigned comment added by 184.108.40.206 (talk) 21:03, 11 August 2011 (UTC)
- The vast majority of musicians, even the most famous, do not have it. But I agree it would be good to have some examples if anyone has a citeable source. Speaking from personal experience (which can't go in the article), when I taught classes for music majors in a large well-known music school, maybe one person out of fifty had it, and they generally were the most talented; however they were outnumbered by super-talented people who did not have perfect pitch. As for specific examples, my Oxford Companion to Music article on absolute pitch lists Wagner and Schumann as two 19th-century composers who did not have it. Antandrus (talk) 21:10, 11 August 2011 (UTC)
- I disagree. This article does not provide a "list" of people who had absolute pitch. If you look waaaay back far enough in the history of this article, you will find just such a list, but that list was removed as trivia. What the unsigned writer above appears to be referring to is the section "Correlation with Musical Talent", which names four specific artists. The point in their being mentioned is not to provide a random assortment of names, but to be representative contemporaries of three specific historical eras. This article already specifies that AP is not required for a musician to be "skilled". If a list of random names is to be generated to make this observation concrete, it would be necessary first to specify the contextual criteria for their inclusion. In other words, for this article, it is not relevant whether any particular celebrity did or did not have AP. If, however, a certain person's possession or non-possession of the ability is/was of encyclopedic or historical note, then there would be an argument for its inclusion. aruffo (talk) 23:27, 11 August 2011 (UTC)
This is a response to Seokhun. The unsourced statements have now been sourced, clarified, and supported, as requested. As Levitin (2008) indicates, there is no evidence, in any published form, that actual AP ability can be learned. Rather, any adult who tries can learn to name some notes, after a few weeks, irrespective of the "method" employed, and this "pseudo-AP" is not comparable to absolute pitch ability (see Meyer, 1899; Maryon, 1916; Brady, 1970; Cuddy, 1970; Nering, 1991; Rush, 1989; Crozier, 1997; Russo, Windell, & Cuddy, 2003; Miyazaki, 2006). aruffo (talk) 22:03, 12 February 2013 (UTC)
- Well, there is a person named David Lucas Burge who claims to have learned Absolute/Perfect Pitch during the ninth grade. His entire story can be found here: http://www.perfectpitch.com/chapter1.htm . I believe the story is genuine, and that is confirmed by feedback from people who have tried his Perfect Pitch course: http://www.perfectpitch.com/success.htm . Now, you could argue that David was not an adult when he learned Perfect Pitch, but I'm pretty sure that at least some, if not several, of those people who have given feedback to his Perfect Pitch course are/were adults. And, it's generally thought that Perfect Pitch can only be learned during the first three-four years of an individual's life; David was not that young. I'm on Seokhun's side. See also http://www.perfectpitch.com/research.htm . HeyMid (contribs) 10:10, 13 February 2013 (UTC)
- Yes. David Lucas Burge has made rather strident claims in his advertising materials. Those claims have not been supported. The two theses that Burge presents as having "proved" his claims, which you have referred to here, not only fail to do so, but actually provide evidence to the contrary (see earlier discussion on this same page). aruffo (talk) 19:34, 13 February 2013 (UTC)
The section on Pseudo AP really needs to be rewritten (or even removed.) The statement "there are no reported cases of an adult obtaining absolute pitch ability through musical training; adults who possess relative pitch, but who do not already have absolute pitch, can learn "pseudo-absolute pitch", and become able to identify notes in a way that superficially resembles absolute pitch" contradicts the earlier statement regarding no external standard on what AP actually means and sounds like a "no real Englishman" argument. Identification of pitch-class is a simple matter of memory and recall; I don't see why stories such as Burge's are treated as sensational or controversial. In the Nature vs Nurture section it is written "all adults who have undergone AP training have failed, when formally tested, to show "an unqualified level of accuracy... comparable to that of AP possessors." Who determines what level of accuracy is necessary to qualify as "real" AP? There is no mention of any quantitative distinction because "real" and "pseudo" AP. Clarissimus (talk) 01:29, 18 April 2013 (UTC)
- The words "external standard" don't refer to the definition of absolute pitch, but rather to an "external" reference tone. The word "external" is actually unnecessary, because there are studies testing whether absolute listeners use internal reference tones, and they don't. I just made a quick grammatical edit to the article which I hope will clarify that matter... I'll write more here in a moment. aruffo (talk) 08:32, 18 April 2013 (UTC)
- I also just flipped the other sentence around, because it did say the wrong thing. As written, as you quoted it, it implied that all adults who have undergone AP training have been formally tested-- which of course they have not. Rather, when formal testing has been done, all adults tested have failed to demonstrate AP ability. You ask who determines the necessary level of accuracy; the referenced article (Takeuchi & Hulse, 1993) is a review article that will point you to the accuracy tests that have been done, but I think you'll find the best measurements in Miyazaki's work. I believe that it was he who came up with the term "pseudo AP"... but even then, he used it differently than is used in this Wiki article. Miyazaki observed that "real" absolute listeners identified tones with almost-perfect and lightning-quick accuracy. Another group of Miyazaki's listeners were slower and less certain in their judgments, and Miyazaki implemented a cutoff of 90% accuracy for these "pseudo" absolute listeners. But even this is greater than the "pseudo" absolute-pitch skill that has been gained by adults who trained to learn it. I don't have the exact figures to mind right now, but I seem to recall that accuracy tended to max out at 25 or 30 percent, at best, and only for a rare few, besides. Furthermore, those tests which were performed were solely of whether individual notes could be named, separately from any musical context, and did not test any other skills (such as those mentioned at the beginning of the Wiki article). In short, the difference between the demonstrated skill of "born-with" AP possessors and the measured results of those who attempted to learn AP is not a close call. aruffo (talk) 08:57, 18 April 2013 (UTC)
- Studies have shown that perfect pitch may not be so attributed to genetics, but may rather be attributed to the use of a tonal language. Is it possible to fluently learn speaking a tonal language as an adult, if you haven't learned one already? HeyMid (contribs) 10:47, 16 February 2013 (UTC)
Linguistics - East Asia
The use of the term "East Asia" is totally wrong in that passage.
1. Vietnam is Southeast Asia, not East Asia. The article makes people think Vietnamese is linguistically related to Chinese because both countries are in "East Asia", which is not the case (both claims).
2. Japanese and Korean are rather pitch-accent languages than real tonal languages. The distinction is made in the penultimate sentence, but completely ignored in the rest of the passage because "East Asia" seems to be China and Vietnam. --220.127.116.11 (talk) 14:39, 14 February 2015 (UTC)
- In the context of dividing Asia into E-W, then Viet-Nam is most definitely "east Asia." It is only when using a N-E-S-W division that "south-east Asia" is the proper term. The article neither intends to show that Chinese and Vietnamese are directly related, nor does it. The linked terms will take the reader to the appropriate language article and there won't be any confusion, Anon IP. HammerFilmFan (talk) 21:53, 11 July 2015 (UTC)