[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Absolute pitch discussion

Dear Robert, Diana, Henkjan, Roy, and others,

Robert Zatorre asked (concerning the recent PNAS study):

"Could it be, therefore, that the over-representation of AP in the
distribution represents a sampling bias?"

Yes, of course. But the crucial question is: Could the bimodal
distribution (clear APers vs. clear non-APers) be the result of a sampling
bias? And here the answer definitely is "No". There is nothing in the design
of the web study that could have attracted zero-APers on the one hand and
excellent APers on the other hand, but then could have repelled moderate and
low performance APers.

Thus the bimodal distribution, as shown in Fig. 1 of the study, has to be
regarded as an empirical fact. The consequences of this finding are
far-reaching. All human traits that are of purely environmental origin only
show one mode in the distribution of its strength.

If a trait shows two modes in the distribution, as is the case with many
pathological traits, we can be certain that a relatively simple genetic
factor is involved. In conclusion, after this new PNAS study, we are now
compelled to expect that a relatively simple genetic factor is involved in
the development of the cognitive version of absolute pitch (AP).

This does not mean that the environmental factors that were mentioned in
this discussion have become less important. But it is now clear that
environmental factors are not sufficient for the development of the AP

The crucial Fig. 1 of the study is easily available here:



Martin Braun
Neuroscience of Music
S-671 95 Klässbol
web site: http://w1.570.telia.com/~u57011259/index.htm

----- Original Message ----- From: "Robert Zatorre" <robert.zatorre@xxxxxxxxx>
To: <AUDITORY@xxxxxxxxxxxxxxx>
Sent: Wednesday, August 29, 2007 6:21 PM
Subject: Re: Absolute pitch discussion

I would agree with Roy's point that the whole issue of AP is complicated, so let's not look for simple solutions. The paper by Deutsch et al is very interesting, but note that its results can be used just as well as evidence either for genetics or for early tone-language exposure, since the higher incidence of AP was found in ethnically Chinese students as compared with non-Chinese; so they were comparing two populations with different genomes as well as different language histories.

As for the PNAS study, it also presents very interesting and valuable
data, and I like the web-based approach they used. However, I wonder if
the two subpopulations seen in the distribution are somehow a
consequence of the recruiting system used. People were not entered into
the test randomly, rather they self-selected to participate, and people
who know they have AP would naturally want to test themselves to see how
they score. Could it be, therefore, that the over-representation of AP
in the distribution represents a sampling bias?


Robert J. Zatorre, Ph.D.
Montreal Neurological Institute
3801 University St.
Montreal, QC Canada H3A 2B4
phone: 1-514-398-8903
fax: 1-514-398-1338
e-mail: robert.zatorre@xxxxxxxxx
web site: www.zlab.mcgill.ca