While I do not wish to speak for Eliot Handelman, a number of years ago (perhaps 5 or more) in a discussion with him the topic turned to a series of melodic dictations, sight-singing and theory exercises I had prepared. He asked me something like "What is this thing with melodic contour?", and went on to, as I understood it, indicate that he did not think that 'contour analysis' was necessarily valid. He will clarify what he said and what he meant. At the time, I thought he either had very poor hearing (which did not seem to be the case), or perhaps he had residual elements of AP, perhaps (as I thought) from taking violin lessons (or piano lessons) at a very young age.
Just before that, I had had the experience with a number of oriental students in my classes who exhibited an unusual (to me) form of AP. They were fine naming notes in the C and G pentatonic scales, and the intervals associated with them, that is, all intervals except the semitone and tritone. In these cases, there was confusion in interval identification between them. To me, one was "close", and the other "more distant". I pondered.
When they sang melodies, in general, the pentatonic notes of C and G, were well-placed and in tune. The other notes, (context dependent) were placed 'somewhere between' these points of reference -- they were 'variable' (or lost) pitches. Since then, I have equated this to having a clear pitch grid for certain notes, and the others being "somewhere between".
Leon's email indicates that there is confusion of pitches which are a third and a fourth apart. My experience in this is that people with good relative pitch may make (contextual) semitone errors, or confuse P4 and P5. I had mentioned before the heretical view that AP individuals do not hear intervals and chords (as integrated) in the same way that relative pitch individuals do. I'm not sure that we live in the same perceptual world.
Best Kevin
Date: Mon, 31 Aug 2009 22:07:46 +0200 From: Leon van Noorden <leonvannoorden@xxxxxxx> Subject: Re: Perception as memory ...
Long ago I have also performed some reaction time studies on note identification. With me as the only subject. Under pressure I made quite a lot confusion errors between fa and la and between si and mi.So I think that AP subjects use a connection between the chroma and the tone height dimensions of pitch.