Re: absolute frequency/absolute color (Eckard Blumschein )

Subject: Re: absolute frequency/absolute color
From:    Eckard Blumschein  <Eckard.Blumschein(at)E-TECHNIK.UNI-MAGDEBURG.DE>
Date:    Tue, 8 May 2001 09:01:59 +0200

Would it be possible to conclude that at least for human the learned absolute pitch includes more than just one element, and it is not necessarily based on a learned tonotopical match? I tentatively imagine relative pitch based on firstly a perhaps innate ability to judge pitch-hight with respect to direction of change and a universal vague measure of any quantity. Secondly there might be some kind of short-time memory of a grid pattern, mediating pitch color by detection of the degree of coincidence with the memorized pattern. Those who are claiming absolute pitch might in addition more or less precisely recall a permanently learned reference grid pattern spanning one octave and derive pitch color from it. Couldn't this explain what I tend to see in the data by Klein, Coles and Donchin in Science 223, 1984: 1307? 1) There are gradual differences between AP listeners and members of a control group. 2) Both the poorest AP listener and the best listener out of the control group show a characteristic pattern of response: 3) Frequency of erroneous responses as a function of the error decreases periodically. It is maximal at and near to whole octaves while it goes through a minimum each time in between. 4) The range of errors extends asymmetrically. It covers about 1.5 negative octave but 2.5 positive octaves for normal listeners. Are there objections or alternative interpretations? Stewart H. Hulse provided additional food for thought. In the book 'The Comparative Psychology of Audition' by Dooling and Hulse, he dealt with 'Comparative Psychology and Pitch Pattern Perception in Songbirds'. He found that songbirds apparently process complex acoustic patterns using both relative and absolute pitch. He also realized a striking frequency range constraint for the ability to discriminate rising/falling frequency. Starlings lost the discrimination when just two initial tones of a probe pattern were outside a trained range. Eckard Blumschein Diana Deutsch wrote: >Actually you only need 12 categories for AP -at least for Western >tonal music - because of octave equivalence. Perception of >differences in height between notes of the same pitch class are not >related to AP. >Christian Kaernbach wrote: >> An absolute memory of 12 colors corresponds to an absolute memory of >> pitch for 12 categories spanning the entire range of audible frequencies >> from 20-16.000 Hz (at my age). I suppose this can be learned in a few >> minutes. A hemitonic absolute pitch memory covering a musical relevant >> range of frequencies would correspond to an absolute memory of roughly >> 90 color shades (only hue, same saturation and brightness). >>Martin Braun wrote: >>> ...this is not possible. There are millions of colors. Most of us, >>> however, would have an absolute memory of 12 colors. Color circles >>> of 12 colors are quite common and can be learned in a few minutes. >>> The 12 tones of our octave can not be learned in years (with >>> extremely rare exceptions), once you are older than 5-7 years

This message came from the mail archive
maintained by:
DAn Ellis <>
Electrical Engineering Dept., Columbia University