New "alpha" version

Comments and questions about AP Avenue.
Post Reply
etaxier
Posts: 136
Joined: Tue Oct 04, 2005 1:57 am
Location: NYC

New "alpha" version

Post by etaxier » Mon Feb 20, 2006 3:26 pm

I'm curious about why you made the (feb. 20) changes. I can't think of any "for" or "against" scenarios, so it would be interesting to see what you're thinking.

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

Post by aruffo » Mon Feb 20, 2006 4:00 pm

I got some splainin to do, eh? Heh.

While I was playing the game, the C-chroma seemed so clear that I wondered, what if ETC/APB has successfully taught me to listen for chroma (with MIDI piano sounds) in general, not just for C? The process isn't designed to teach memory for pitch, anyway-- just chroma perception-- so why would it be important to do only one pitch at a time?

Following that, I considered the magic number seven (plus or minus two) and how, theoretically, that would allow our categorical perception (with or without memory factors) to ably handle all seven pitches of the C-major scale simultaneously.

Then I noticed my own experience playing with the G-pitch; very frequently I would wonder if I'd heard a G or not-- not because I failed to recognize a G, but because I didn't have anything except C to compare it to that would assure me that it wasn't G ("no, I can tell it's not a G, because it's that other one instead").

In a standard context, simply comparing one tone to another leads to tonal recognition instead of chroma perception, but if the C-pitch had done its work then the chroma would already be evident, and the player would be listening for that instead of tonal quality. I further reasoned that any issues of listening to scale degree instead of chroma would be eliminated in later levels, especially with the re-introduction of octaves.

Therefore, it seemed reasonable to introduce all six new pitches at once, provided that the blue buttons get filled up with comparison tones.

So far, it seems to be working to the extent that I am occasionally recognizing them without resorting either to the blue buttons or to their relationship to each other (other than "it's not that one, it's this one" which is actually the goal). But I haven't yet figured out the best way to advance the player other than inching each pitch forward with successful recognitions, and that's pretty slow going. Although it's still fun to measure waves of aliens, the rank advancement has slowed to almost nothing (since it's spread out across seven pitches).

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

Post by aruffo » Tue Feb 21, 2006 5:23 pm

So far, it's a fairly unsatisfying experience. I'm not sure whether this is because the game mechanics were designed for one-pitch-at-a-time or if there is some specific detriment introduced by having all seven pitches or if I just need to keep at it longer. I think I'll puzzle out the theory of it a bit more before I modify the game design any further-- I don't want to take a big step in an unproductive direction which could be avoided with a little thought-work.

etaxier
Posts: 136
Joined: Tue Oct 04, 2005 1:57 am
Location: NYC

Post by etaxier » Wed Feb 22, 2006 4:15 am

I don't know why, but I'm inclined to ask a bunch of questions that you've probably already answered before. Still, I hope it might help you work through it, and I hope it might help me understand what you're working through!

How can you learn to listen to pitch chroma?

How do you know you're listening to pitch chroma and not some other phenomenon?

Does it really only take one pitch in one octave to learn to listen to pitch chroma?

How does perception of pitch chroma interact with memory?

What role does expectation play? Typically, "musical expectation" applies to motifs, phrases, melodies, gestures, etc., and not to individual notes; if expectation of individual notes is a goal, should the before/after sequence of flag/sound be effected?

Do you have to learn to ignore anything in order to learn to hear pitch chroma?

What's implied if the game gets much more difficult whenever new notes are added? Does it mean players have to "re-learn" pitch chroma whenever they're confronted with a new note, or does it mean that players are gradually being forced to compare notes instead of memorize them, or does it mean that players never compare notes and always use short term memory...?

What's the difference -- if there is one -- between recognizing something by memory and recognizing something by perception? If there is a difference, which one does ETC teach, and how does it exclude the other? If there isn't a difference, then how does ETC uniquely incorporate long term (or "automatic") memory?

How can ETC bridge the gap between recognizing a pitch in a melody through eidetic comparison, and recognizing a pitch in a novel melody (including improvisation)?

How do the answers to these questions directly influence the mechanics of the game?



That's about all I can think of for now. Good luck, and sorry for the barrage!

Eric

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

Post by aruffo » Thu Feb 23, 2006 12:26 am

Your questions are as good a place to start as any-- even if I have answered them before, the challenge will be to restate that answer as briefly as possible. I was idly thinking about it today, but hadn't found the right foothold as yet.

How can you learn to listen to pitch chroma?


Scroll down to August 15 to see how it works.

How do you know you're listening to pitch chroma and not some other phenomenon?


I'm somewhat disenchanted with the term "pitch chroma" because the two words have come to mean practically the same thing to me. Specifically: pitch, or chroma, is the psychological response to the vibratory frequency of an energetic impulse. The only reason that the term "chroma" exists is because "pitch" is so frequently confused with other tonal characteristics.

Anyway, I do have an answer, although it seems facetious: you know you're listening to pitch (chroma) because it can't be anything else. To that end:

Does it really only take one pitch in one octave to learn to listen to pitch chroma?


Yes and no. It's clear that a pitch quality will attach itself to other tonal characteristics so that tones in different octaves, on different instruments, at different volumes, etc, will seem to be different pitches. So no, you can't learn total chroma perception until and unless you have been listening for pitch in all its disguises. But yes, according to the theory, the reports presented in this forum, and my own experience, one pitch in one octave will indeed force even a naive listener to hear the unique quality ("chroma") of a pitch.

How does perception of pitch chroma interact with memory?


I still haven't researched this, but here's my current understanding: once you recognize something as a unique concept, you can remember it. If you can create a paired associate (such as a grapheme) you can use that associated thing as a shorthand reference to the otherwise self-referential concept, thereby "fixing" the memory of that concept.

What role does expectation play?


I don't know. I suspect the role is gigantic, and would probably be parallel to any available research on language or optical tricks. At this point, the only explicit use of expectation in APB is that the player must be aware that the target pitch may not be present; if they knew there would always be one or the other it'd be too easy to ascribe imagined qualities to the sounds.

Do you have to learn to ignore anything in order to learn to hear pitch chroma?


I'd say you have to dismiss things, but you probably don't have to ignore anything.

What's implied if the game gets much more difficult whenever new notes are added?


Does it get much more difficult when new notes are added? I can imagine some reasons why it might get harder, and other reasons why it might get easier. If it does get more difficult, I'd probably want to know why the players feel it becomes more difficult and speculate about their observations.

If it were to get more difficult, I suspect it would be because there are new comparisons to be made. Where the player's mind has been steadily excluding and eliminating irrelevant sense data to focus in on a restricted set of signals, now they must again detect what's unique about this new signal-- which means they have to evaluate not merely the new pitch for what it is, but all eleven of the other pitches for what the new one isn't. A similar effect occurs in Interval Loader, I've noticed-- when I get a new interval, I start second-guessing the ones I so easily recognized before that point.

What's the difference -- if there is one -- between recognizing something by memory and recognizing something by perception?


A kind of answer: which parts of the brain are activated. I read a study about memory for melody; they did brain scans, asking subjects to hear a melody and then to imagine the same melody from memory. The scans showed that the same parts of the brain were being used in both cases-- except that, as you might expect, the remembered melody did not use the parts of the brain involved in receiving and processing the sensory data.

In either case, I believe that the important goal is to form concepts for pitch sounds so they can be remembered. That's what the melody word will do.

Check out Robert Goldstone's article about perceptual learning, if you haven't already.

How can ETC bridge the gap between recognizing a pitch in a melody through eidetic comparison, and recognizing a pitch in a novel melody (including improvisation)?


I notice that you say how can ETC do it, rather than how does ETC do it-- which is good, because ETC doesn't do it yet. The answer is, I'm convinced, through structural comprehension and literacy training-- I'm convinced of this by the fact that people who already have stronger structural knowledge of music find the APB training has a greater effect on their musicianship.

How do the answers to these questions directly influence the mechanics of the game?


Well, I think in the short term they may have persuaded me to stick with one-pitch-at-a-time. The reason I think I've been finding the multiple-pitch-at-a-time gameplay so unsatisfying isn't because of the rank issues (I'm very patient) but because there seems to be little motivation to form a mental concept of the pitch. Although it is entirely possible that, if I were to continue to play, the mental concepts would form unconsciously, the fact is that even now I'm listening for C and G as though they were familiar friends and for all the others I'm relying almost totally on the blue buttons and not making much of an effort to recognize the pitch.

Still, I sometimes do recognize the pitch, so I'll keep playing it this way for a while longer to see if I start to form those conceptual categories-- but even if theoretically I would eventually achieve results this way, it's just not as satisfying as being so quickly able to pick out and remember a single new pitch.

etaxier
Posts: 136
Joined: Tue Oct 04, 2005 1:57 am
Location: NYC

Post by etaxier » Sat Feb 25, 2006 5:08 am

Thanks for the reply. I did read the Goldstone paper, as well as some more of his ideas (I like his emphasis on concept<--->perception feedback, plus his model of unitization and segmentation, which reminds me of Hegelian dialectic).

I also found a concentrated paper by D.J. Levitin, at [edit: link and summary available at "phase 11" 1999 -- see below]

I should also like to correct your summary of Ron Gorrow's "Hearing and Writing Music." I read through it some time ago, and it's primary goal is to teach musical transcription. It does include "distance between pitches," but only as one optional technique out of several. His theory is to encourage you to use a variety of mnemonic devices, and eventually acquire (through practice) a hollistic and efficient transcription method. The major failing of the book is that it leaves all of the actual learning up to the reader's ability to develop his own exercises. I can borrow a copy, if you like, and get you more details, since you may find portions of it useful for the "literacy" training aspects of ETC.
Last edited by etaxier on Mon Feb 27, 2006 2:09 am, edited 1 time in total.

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

Post by aruffo » Sat Feb 25, 2006 12:19 pm

Ah! Thanks for that tip about Gorow's book; I'll add your comments to the "library" page unless you have an objection to that. I used to have the book on my shelf-- I've recently been disposing of things that I can find at libraries because my stint at this university is almost done-- and perhaps I will take another look at it later.

If you ever wonder what papers I have or haven't seen, just look at Phase 11 to see if you find it mentioned there.

Post Reply