Acoustic Learning, Inc.
Absolute Pitch research, ear training and more


 

"The value of learning music is not in the number of pieces one may play, but in the musical thoughts one can think."

 - Evelyn Fletcher-Copp (1872-1944)

Hello!  I'm Chris Aruffo, your host here at Acoustic Learning.  It's my goal to dispel the myth and the mystery that surrounds the phenomenon of perfect pitch (also known as absolute pitch), and to make it a skill accessible to anyone who wants to learn it.

To explore ways you can learn absolute pitch, click "Learn AP" or the research links above.


Recent Discussion


April 10 – Fixing a hole

I've been playing Level 4 now for a while.  Moving through the Small and Medium games was unexpectedly easy.  Only the two new categories were featured (green egg and orange egg) and it seemed blindingly obvious which category each tone belonged to.  I even got two perfect scores on the Small game without much effort.  This made it all the more perplexing when I started in on the Large game, where the three remaining categories were re-introduced, and suddenly I no longer had this same clarity.  What's more, my perception of the lowest category (blue-grey egg) was still very weak.  I decided to take a moment and figure out why I was making these mistakes, because by this time I felt I should know better.

The answer was height.  I had come to think of blue-grey egg as "the lowest" category-- therefore, when I heard a tone that wasn't "the lowest", I'd quickly assume that it must be the next-highest category (brown egg), and answer incorrectly.  Likewise, on the opposite end, I thought of orange egg as "the highest" category, and when I heard a tone that wasn't "the highest", I'd suppose it must be the next-lowest category (green egg), and answer incorrectly.  And then, thirdly, because I could hear a very distinct "plunky" quality from the lower tones of the red-yellow egg category, every time I heard a tone I could tell was in the upper half of the overall range, but lower than the highest category orange egg, I would guess green egg unless I distinctly heard that "plunky" quality... and only after seeing that I'd gotten it wrong did I listen more carefully and hear that same "plunky" quality more subtly present.  Once I paid attention, it was evident that I was making these same mistakes constantly, and always because of the height judgment.

I brought up the pitch-stats screen and saw quite clearly what height errors look like.  They curve down toward the category boundary, because as a pitch moves toward the border, the pitch increasingly seems "high enough" or "low enough" to be part of the next category.  I've marked them in red, below, to make them even more obvious.

height stats

I noticed, of course, that the other boundaries had the sharp drop-off indicative of categorical perception (that is, low brown egg, the division between brown egg and red-yellow egg, and either side of green egg).  And I quickly realized that my subjective experience of these borders was different.  I have a clear sense of what brown egg, green egg, and the lower part of red-yellow egg sound like.  Each of these areas has a characteristic sound.  As I was playing the game, tones within these areas didn't really lose their characteristic sound until they got to the border.  Once they passed the border, then they showed more of the other category's characteristic sound.  As "abminor" described it in the forum, "[my perception] tells me this is like a color blend between the two categories but one color dominates, so I select the appropriate category based off that."  Where I knew the characteristic sounds, I showed categorical judgment, and didn't make height errors.

So I no longer have to wonder whether the categorical perception being taught here is based on height.  It can't be, because now I see that you can't form categories if you make judgments based on height.  In fact, it suddenly seems obvious why and how chroma and categorical perception work together in absolute pitch judgment, which in turn explains why I was so baffled about how to learn categorical perception along a single dimensional continuum:  that is, you can't.  The "height" of a tone doesn't change from end to end.  You can tell that one tone is higher or lower than another, and you can estimate approximately where a tone sits along the entire continuum of frequency values, but because "height" is a quality that can be detected along the entire range of sound, there is no reason why your mind should mark a particular frequency x as the divisor between any one area and any other area.  So your mind doesn't do that.  And it won't do that, no matter how much effort you put into training.  But once your mind detects some quality that actually does change-- fundamentally, not incrementally-- from one area to another, then it becomes possible for your mind to mark a division between those changes.  Chroma is an obvious candidate for that quality.  So are scale degrees, for that matter.  But height isn't.

With that in mind, I've altered my gameplay strategy.  Because I know, now, that I have terrible trouble mistaking blue-grey egg for brown egg, and the parts immediately above and below green egg for green egg, I no longer make it my goal to place these eggs into their respective bins.  Rather, when I hear a tone that could belong to blue-grey egg, brown egg, or green egg, I make a mental guess at what category it belongs to and then set it aside.  Gradually I build a big enough pile of eggs from a particular area that I can start sifting through them and finding the boundary, so that I can sort each of them into their respective bins without making mistakes.  Being able to do this, by the way, is why the game is structured so you can randomly access the tones, instead of being presented with them one-at-a-time quiz-style.  You need to be able to make comparisons, hear the similar qualities between the tones, and make judgments based on their similarities.  Once you have formed an idea in your mind of what each category sounds like, you can compare the sounds you hear to that image in your mind, but until that image is clearly and solidly formed there's no other way to know except by comparing the sounds to each other.

In addition to all this, I had a sudden flash of insight about what it means to have "active" absolute pitch-- to be able to sing whatever pitch you want on demand.  Having active absolute pitch means that you do not sing a pitch.  You sing the quality that you associate with the pitch, and the correct pitch comes out automatically.  Compare this to how you produce language sounds.  You don't torture yourself trying to remember what the frequencies 240 Hz and 2400 Hz sound like together and then trying to make yourself reproduce those pitches.  You just think "ee" and those frequencies are what come out.  Likewise, I just now tested myself by singing the "plunky" quality I associate with the lower part of the red/yellow category.  This happens to be middle C-- but I just sang it as a quality, without thinking of a melody, without thinking of a height, without even thinking of it as a pitch.  Just remembered the quality, opened my mouth, and let it out.  And wouldn't you know?  It was a C.  Not a perfectly-tuned C, no, because the category I'm learning is still rather broad-- but it was a middle C.  This, in turn, answers the question I've been beating myself up about for years:  when you're trying to identify or produce a pitch, how do you know that you've got it right?  How can you confirm your pitch is correct, except by playing a "reference tone" and finding out?  The answer is finally clear:  because it matches the sound quality you've got in your mind.  The sound quality is independent of height-- it's particular only to the pitch category that you're trying to identify (or produce).  Therefore, once you have learned the sound quality of that category, then you will be able to identify (or produce) the tone, because no other category has that same quality.  The critical element to success is whether the quality you've learned is in fact that essential quality of the category.  If you have learned to estimate height, it will not work, because all pitch categories have height.  If you have learned by trying to apply adjectives to pitches (mellow, twangy, etc) it won't work, because you become able to describe pitch qualities by being able to recognize pitches, and not the other way around.  But once you can detect and distinctly recognize the sound quality that uniquely defines a pitch category... then, you can identify or produce a tone, and you'll know that you've done it right.


"Compared to what we ought to be, we are only half awake. We are making use of only a small part of our physical and mental resources. Stating the thing broadly, the human individual thus lives far within his limits. He possesses powers of various sorts which he habitually fails to use."

- William James  (1842-1910)


Home