Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rutgers!princeton!mind!harnad
From: harnad@mind.UUCP (Stevan Harnad)
Newsgroups: sci.med
Subject: Re: tone deafness?
Message-ID: <210@mind.UUCP>
Date: Tue, 11-Nov-86 11:58:56 EST
Article-I.D.: mind.210
Posted: Tue Nov 11 11:58:56 1986
Date-Received: Tue, 11-Nov-86 20:20:43 EST
References: <2376@bu-cs.bu-cs.BU.EDU> <3808@columbia.UUCP>
Distribution: net
Organization: Cognitive Science, Princeton University
Lines: 51
Summary: Definition of Absolute Pitch

In article <3808@columbia.UUCP>, zdenek@heathcliff.columbia.edu (Zdenek Radouch) writes:
> 
>    Given that we don't completely understand the mechanism of hearing
> and that we can only guess, how the information is processed in the brain,
> I'm really surprised to hear that there is an "evidence" that absolute pitch
> (perfect pitch) is inherited. What's that evidence?...Also note that there
> is no explicit definition of what "absolute pitch" is.

Absolute pitch is a special case of what psychophysicists call
"absolute judgment" (also called absolute discrimination,
identification, categorization, labeling). This is usually contrasted
with relative judgment (or relative disctimination, or just
discrimination). In relative discrimination, pairs of stimuli are
presented, and the subject must perform a relative comparison, usually
a same/different judgement, or a degree-of-similarity match. In
identification, on the other hand, the stimulus is presented alone,
and it must be given its correct (arbitrary, learned) "label" or name.

George Miller wrote a famous paper in Psych Review in 1956 about the
informational limits on relative vs absolute judgments ("The Magical
Number 7 +/- 2"). Identification was governed by the 7-chunk
constraint, discrimination varied with the sense modality in question
and with the subject's sensory acuity. Identification capacity could be
increased by "rechunking" or recoding the stimuli in question. The
modern incarnation of this area of research is called "categorical
perception," and I'm editing a book by that name which will appear in
April, published by Cambridge University Press. (An active discussion
on aspects of it is now going on in net.ai under the heading: "Searle,
Turing, Symbols, Categories.")

Absolute pitch (AP) is discussed in the book. Note that what most people
call "relative pitch" is in fact a short-term instance of absolute
judgment. The idea is that people with AP can identify all the
semitones (C, C#, D, etc.) absolutely, when presented in isolation.
Those with "relative pitch" can do the same thing, but only for a
short period, while an identified "reference tone" is still fresh in
their ears.

Empirical questions about AP (and categorical perception in general),
include whether it is innate or learned, what role physical, sensory,
motor and cognitive constraints play in it, and what is the nature of
the underlying representation. (Other examples include color
identification and phoneme identification.) I presume that "tone
deafness" refers either to absence of relative pitch or to radically
diminished discriminative acuity. As long as someone is not entirely
deaf, some frequency discrimination must be present.
-- 

Stevan Harnad                                  (609) - 921 7771
{allegra, bellcore, seismo, rutgers, packard}  !princeton!mind!harnad
harnad%mind@princeton.csnet