Why DON'T people develop perfect pitch?

Thoughts and responses regarding the research at acousticlearning.com.
lorelei
Posts: 221
Joined: Sat Mar 20, 2010 7:36 am

Post by lorelei » Mon Mar 22, 2010 2:32 pm

Notes are named by alphabet letters, but they don't behave like letters in a piece of writing in that aspect. Rather, they make up certain melodic phrases in pieces, and indicate what key a piece is in. The keys in pieces correspond to mood and overall color scheme in a painting (I sometimes compare music to visual art because it's the closest thing you can get and most people can see colors like I hear pitches.), and the melodic phrases correspond to little details that are nevertheless crucial in the overall painting.
When I'm listening to a piece of music, I do perceive the "direction," but this is not the most important thing. The most important thing is the flow of the melody, and where the melody leads you.

aruffo
Site Admin
Posts: 1692
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Post by aruffo » Mon Mar 22, 2010 3:42 pm

That's curious... is my two-part question nonsensical? I may get a better answer by not trying to put words in your mouth! Namely...

When you say you "perceive the 'direction'," how would you describe that experience (physically and cognitively)?

lorelei
Posts: 221
Joined: Sat Mar 20, 2010 7:36 am

Post by lorelei » Tue Mar 23, 2010 2:48 am

What I meant, is if there are just two notes, such as C and D, eg. an ambulance siren fading into the distance, then I jsut hear a C and D. However, in a piece of music, I don't pay attention to just the notes. It's a bit like listening to poetry, since you don't pay attention to every letter that goes by, you listen to the overall scheme of things. Sorry, this is kinda hard to describe, cause I can't imagine what listening to music without AP would be like, and I've never experienced it.
By the way, when you said that music to a person without AP sounds like humming, what did you mean, since even humming has a clear pitch and melodic flow to it (unless it's out of tune and irritating)?

aruffo
Site Admin
Posts: 1692
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Post by aruffo » Tue Mar 23, 2010 10:58 am

It strikes me that in no Alice In Wonderland adaptation that I've ever seen has the actress portraying Alice spoken the line "Curiouser and curiouser" in a way that sounds at all natural. (I fault the director.)

It continues to be telling that I don't get "yes" or "no" answers to what I believe to be yes-or-no questions... thanks for taking a stab at it, if only because it continues to make me think that I'm right, and certainly confirms that if I want to be sure I'm right, indirect testing-- and not direct questioning-- is the only way to be even remotely sure. It would take extensive discussion just to make sure we're talking about the same things, especially if we use the same terms!

Changing the topic somewhat: I've recently been tasked to come up with some decent tests of musical ability which incorporate absolute listening. That is, there are ways to evaluate "musical skill" (going back to Seashore and his ilk) but none of them actually take into account any level of absolute-listening skill. It must be possible to develop a test which will not simply determine whether or not someone "has" absolute listening skills, but to what extent this ability exists and what manner of influence it exerts on musical production and conception.

lorelei
Posts: 221
Joined: Sat Mar 20, 2010 7:36 am

Post by lorelei » Fri Mar 26, 2010 2:51 am

Hmm... a theory for a test of how each person hears music as could be:
-place a color wheel or one of those silly emotions magnets in front of the test subject and tell them to show which emotion best corresponds to the pitch or piece of music that was played
-play different pitches and/or different pieces and see what the subject thinks of it
Do you think this would prove or show anything?

aruffo
Site Admin
Posts: 1692
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Post by aruffo » Fri Mar 26, 2010 4:29 pm

If the answers were consistent, it might show that the listener has a specific and recognizable concept for each pitch.

lorelei
Posts: 221
Joined: Sat Mar 20, 2010 7:36 am

Post by lorelei » Sat Mar 27, 2010 2:04 am

Hmm... would this test evaluate absolute listening skill? You can try variations: see how small a frequency difference from A=440 Hz annoys the listener. Also, if the results are more consistent, it can be said that their ability is better, or something like that, right?

SteveA
Posts: 23
Joined: Thu Jan 28, 2010 4:43 am
Location: Amsterdam, NL

Post by SteveA » Sat Mar 27, 2010 8:17 am

The way I would test for AP is like this.

I'd create a MIDI file of a reasonably long piece of music, say in A major. The music would be 2 minutes in length, and would stay in A major throughout.

Then I'd modify the MIDI file to create two new MIDI files.

1. The original MIDI file altered so that the pitch gradually increases, cent by cent, throughout the 2 mintes, so that it starts in A major and ends in Bb major.

2. The original MIDI file transposed into Bb major, and cut off after 20 seconds.

Then, to run the test, I'd play the listener the first piece, then play the second piece, and ask "what's happening here?"

I would expect someone without AP to say "it's the same piece starting again".
I would expect someone with AP to say "it's like the first piece, but this one starts on a different pitch".


As a control, I'd also test the original midi file in A major followed by the first 20 seconds of the original midi file in A major.

I would expect anyone to say "it's the same piece starting again".

aruffo
Site Admin
Posts: 1692
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Post by aruffo » Sat Mar 27, 2010 10:03 am

The "consistent answers" is an idea suggested to me by Dan Levitin when I met him at the L.O.V.E. conference in February.

Knowing that absolute processors have categorical perception of pitch frequency, and suspecting that absolute processors do not perceive pitch classes as "higher" or "lower" than each other in magnitude (but rather in nominally ordered sequence), I said it seemed likely that absolute processors had specific and unique mental concepts for each pitch class.

That is, a D would not be a D because it was "more than" C or C#, but because it was... and here's where the AP processor would free-associate. Levitin pointed out that if the pitch-category actually was a mental concept, and not a spatial measurement, then the free association would be consistent over time. Bring the listener back any time, put the pitch into different timbres and key contexts, and when asked to free associate about the characteristics of that pitch, the responses would remain essentially the same.

lorelei
Posts: 221
Joined: Sat Mar 20, 2010 7:36 am

Post by lorelei » Sun Mar 28, 2010 1:01 am

SteveA, I like your idea, and you can make it even more consistent by:
if a listener tells you the second is in a different key than the first, ask them which keys both of them are in.
Aruffo, I think that what you said about the "consistent answers" is correct. If you play me a D, I know it's a D, not because of its' relationship to C but just because it "sounds like" a D. If you play me an A on a keyboard and an A as a pedal tone in a complicated piece, I can tell you the pedal tone is an A.

lorelei
Posts: 221
Joined: Sat Mar 20, 2010 7:36 am

Post by lorelei » Tue Jun 08, 2010 1:59 am

to clear up what i've been saying is, yes, I do perceive height, but chroma is more important. If you compare it to color, you do perceive the brightness of color, but color itself (red, green, blue) is more important. Sorry for making it sound confusing.

Also,
SteveA wrote:1. The original MIDI file altered so that the pitch gradually increases, cent by cent, throughout the 2 mintes, so that it starts in A major and ends in Bb major.

People with AP would get annoyed with the transposing pretty quickly and it would be hard for them to listen to the whole thing. You could already tell then to some extent. Why have the midi file transpose cent by cent?

kimleonard
Posts: 1
Joined: Fri May 11, 2012 12:47 am

Post by kimleonard » Fri May 11, 2012 12:49 am

A knowledge of all instruments; their ranges, how they sound in their different registers, how they sound in combination, what things are not possible on certain instruments (for example, certain trills are not possible on some instruments). A decent sense of relative pitch. The ability to step outside of one's work and assess it critically, and to work and work on it until it becomes something worth hearing.
develop perfect pitch

Post Reply