American Journal of Medical Genetics 98:280 ±282 (2000)

Letter to the Editor

Early Childhood Music Education and
Predisposition to Absolute Pitch:
Teasing Apart Genes and Environment

To the Editor: education are designed with the express purpose of

developing AP ability. These include the Yamaha
Twelve years ago in this journal, Pro®ta and Bidder method, the Royal College method, and other methods
[1988] reported the ®rst evidence for familial aggrega- termed ``®xed do.'' ``Fixed do'' pedagogy expressly
tion of a rare cognitive phenotype known commonly as associates solfeÂge syllables (do-re-mi, etc.) with parti-
``perfect pitch'', also designated ``absolute pitch.'' Since cular standard pitches: for example ``do'' is always C,
that report, several studies [Baharloo et al., 1998; ``sol'' is always G. Other types of training are more
Gregersen et al., 1999] provided evidence for genetic geared to the development of overall musicality, and
and environmental effects in predisposing to the develop- emphasize the development of relative pitch ability,
ment of absolute pitch (AP). Absolute pitch is a relatively with minimal if any attempt to train for absolute pitch
uncommon cognitive ability possessed by a minority of recognition. These include the Suzuki method, and
professional and amateur musicians, characterized by other ``moveable do'' techniques. ``Moveable do'' peda-
their ability to identify pitch names, or to recall speci®c gogy associates solfeÂge syllables with a scalar function
pitches, without bene®t of a reference pitch; this is within a key, so that ``do'' can associate with different
usually accomplished in a relatively effortless and pitches, depending on the key being utilized for the
instantaneous fashion [Ward and Burns, 1982; Greger- training exercise. Instrumental music lessons gener-
sen, 1998]. The prevalence of AP in the general ally place no emphasis on absolute pitch ability, and we
population is unknown because it cannot be ascertained did not categorize them as part of ``®xed do'' pedagogy,
in subjects unless they possess at least a minimum although of course, practice on most instruments does
amount of musical education. The prevalence of AP in reinforce the association of particular note names with
populations of students in professional level music speci®c pitches. We also asked about other forms of
schools or in the personnel of major orchestras is in the exposure to music, such as the presence of a parent or
range of 10±15% [Baharloo et al., 1998; Gregersen et sib who either studied or taught music in the home.
al., 1999]. This is in contrast to the prevalence estimate Respondents were speci®cally asked to indicate which
of 1/1,500 in amateur music students reported pre- of these types of exposure to music they had received
viously [Pro®ta and Bidder, 1988]. A history of formal before the age of 7.
musical training before age 7 is widely acknowledged to The overall rate of AP in this population was 12.2%.
predispose to AP; however, such training is neither Similar to results we reported previously [Gregersen et
necessary nor suf®cient in most persons. As noted above, al., 1999], there was a markedly increased rate of
substantial familial aggregation of AP occurs, although AP among Asian students (42/80; 47.5%) compared
it is unclear to what degree this re¯ects genetic factors with Caucasian students (75/834; 9.0%). The relatively
[Baharloo et al., 1998; Gregersen et al., 1999]. higher rate in Asians was present among all the major
We report the results of a survey of 1067 music ethnic subgroups - Japanese (26% AP‡), Korean (37%
students who were enrolled in music theory classes in AP‡) and Chinese (65% AP‡). One possible explana-
one of 13 educational institutions in the United States. tion for this difference might be that early childhood
In addition to questions concerning AP ability and music exposure is more frequent in Asian students.
family history, we attempted to establish the nature of There was no signi®cant difference, however, between
the musical training that these students received these two ethnic groups, with 80% of Asians and 71% of
before the age of 7. Some types of early childhood Caucasians reporting early music exposure of at least
some type (P ˆ 0.09). When the type of early childhood
*Correspondence to: Peter K. Gregersen, M.D., Division of
music training was compared, however, Asians were
Biology and Human Genetics, North Shore University Hospital, signi®cantly more likely to have been exposed to early
350 Community Drive, Manhasset, NY 11030. E-mail: training based on ``®xed do'' techniques compared with
[email protected] Caucasians (29% vs. 6.0%, P ˆ 0.001).
Received 25 July 2000; Accepted 29 September 2000 We also performed a logistic regression using dif-
Published online 10 January 2001 ferent types of early childhood music training among

ß 2001 Wiley-Liss, Inc.

 Letter to the Editor 281

childhood training substantially enhances the prob-

ability of AP, but is not an absolute requirement for its
development.
The lack of an absolute requirement for early
training is also shown by the fact that among Asians
with AP, 15.8% of our sample reported no early music
exposure or training; this ®gure was 6.7% in Cauca-
sians with AP. In addition, high levels of training do not
guarantee the development of absolute pitch. For
example, among students without AP, 4.7% reported
early childhood training using both ®xed and moveable
do techniques before age 7.
We have previously provided an estimate of ls
(relative risk to sibs) for AP, by comparing the rate of
AP in the sibs of AP‡ and APÿ index cases [Gregersen
et al., 1999] Using this new set of data, we again
Fig. 1. Results of a logistic regression analysis to examine the role of calculate a signi®cant overall lsˆ12.2. This ®gure is
race (Asian vs. non-Asian) and early childhood music exposureon the calculated as follows:
probability of developing AP. Musical exposure/training was divided into
three types: 1) No training before age 7; 2) exposure or training that
predominantly emphasize ``moveable do'' methods and the development of Rate of AP in sibs of music students with AP 15:9%†
relative pitch, and 3) training that emphasizes the development of absolute
pitch skills based on ``®xed do'' methods.
Rate of AP in sibs of music students without AP 1:3%†

Because the denominator in this ls calculation

utilizes the AP rate in sibs of music students without
AP, we have argued previously that this corrects
somewhat for the role of environment [Gregersen et
individuals with and without AP, also taking account of al., 1999]. This is based on the supposition that most
ethnic background (Asian vs. non-Asian). We consid- music students and their sibs have some degree of
ered separately students without any training at all exposure to music when growing up, and are at least
before age 7, those who received training based on more similar to one another in this regard when
``moveable do'' (including instrument instruction or compared to the general population. Clearly, if the
exposure to relatives teaching or studying music in the traditional means of calculating ls were utilized, that
home), and those whose training included ``®xed do'' uses overall population prevalence in the denominator
methods. The results are shown in Figure 1. Note that [Risch, 1990], the resulting ls value would be extremely
for all childhood training environments, the probability high. This is because AP in the general population is
of AP in Asians exceeds that for non-Asians. Further- probably at least an order of magnitude lower than in
more, the predicted AP rate is greater in the setting of sibs of APÿ music students [Pro®ta and Bidder, 1988],
more early training, especially training based on ``®xed and in any case is unmeasurable.
do'' methods. This emphasizes that early childhood The current study allows us at least to partially
music training has a major impact on the development address the role of environment in the contribution to
of AP. Nevertheless, the logistic regression analysis this estimate of ls. Thus, we have also examined the
also predicts that even in the absence of musical rate of AP in various groups of sibs, strati®ed by the
training or exposure before age 7, the predicted amount of early childhood exposure reported by the
probability of Asian music students having AP is 0.20; index cases (either AP‡ or APÿ music students). These
among non-Asians, this probability is 0.03. Thus, early ®gures are shown in Table I for 8 different groups of

TABLE I. Sib Recurrence Rates for AP in Various Subsets of Music Students*

Sib group Sibling population Sibs with AP n (%)

1 Sibs (nˆ182) of AP ‡ music students 29 (15.9)

2 Sibs (nˆ1448) of APÿ music students 19 (1.3)
3 Sibs (nˆ726) of APÿ music students who had training (any type) before age 7 13 (1.8)
4 Sibs (nˆ722) of APÿ music students who did not have training (any type) before age 7 6 (0.8)
5 Sibs (nˆ157) of AP‡ music students who had training (any type) before age 7 25 (15.9)
6 Sibs (nˆ25) of AP‡ music students who did not have training (any type) before age 7 4 (16)
7 Sibs (nˆ35) of AP‡ music students who had ``®xed do'' training before age 7 8 (22.9)
8 Sibs (nˆ86) of APÿ music students who had ``®xed do'' training before age 7 1 (1.2)
*Group 1 vs. Group 2 (P < 0.001); Group 1 vs. Group 3 (P < 0.001); Group 3 vs. Group 4 (PˆNS); Group 5 vs. Group 6 (PˆNS); Group 7 vs. Group 8
(P < 0.001).
282 Gregersen et al.

sibs. Groups 1 and 2 give the overall sib AP rates, as REFERENCES

discussed above. There is no signi®cant difference in Baharloo S, Service SK, Risch N, Gitschier J, Freimer NB. 2000. Familial
the rate of AP in the sibs of APÿ students who had aggregation of absolute pitch. Am J Hum Genet 67:755±758.
(Group 3), or did not have (Group 4) early childhood Baharloo S, Johnston PA, Service SK, Gitschier J, Freimer NB. 1998.
training ( P ˆ 0.11). Furthermore, early childhood Absolute pitch: an approach for identi®cation of genetic and nongenetic
training also does not appear to in¯uence AP rates in components. Am J Hum Genet 62:224±231.
the sibs of propositi with AP (Group 5 vs. Group 6). Gregersen PK. 1998. Instant recognition: the genetics of pitch perception.
Am J Hum Genet 62:221±223.
Perhaps most compelling is the signi®cant difference
Gregersen PK, Kowlasky E, Kohn N, Marvin EW. 1999. Absolute pitch:
between Group 7 and Group 8. These two groups prevalence, ethnic variation, and estimation of the genetic component.
contain sibs of students who all underwent ``®xed do'' Am J Hum Genet 65:911±913.
training before age 7. Note that in Group 7 (sibs of AP‡ Miyazaki K. 1988. Musical pitch identi®cation. Percept Psychophysiol
propositi) the sib recurrence rate for AP is 22.9%. In 44:501±512.
contrast, in Group 8, where the index cases are all APÿ, Pro®ta J, Bidder GT. 1988. Perfect pitch. Am J Med Genet 29:763±771.
the sib rate of AP is only 1.2%. Thus, if early childhood Risch NJ. 1990. Linkage strategies for genetically complex traits. I.
``®xed do'' training in the index cases is treated as a Multilocus models. Am J Hum Genet 46:222±228.
surrogate indicator for a generally predisposing envir- Risch NJ. 2000. Searching for genetic determinants in the new millennium.
Nature 405:847±856.
onment in the family, the data indicate that this has no
Ward WAD, Burns EM. 1982. Absolute pitch. In: Deutsch D, editor. The
impact on the extent of familial aggregation. psychology of music. New York: Academic Press. 431±451.
In summary, we suggest that AP provides a unique
opportunity to investigate the role of gene±environ- Peter K. Gregersen*
ment interactions on a relatively ``clean'' cognitive Elena Kowalsky
phenotype [Miyazaki, 1980; Baharloo et al., 1998]. Division of Biology and Human
The most reasonable view of the existing data is that Genetics,
certain early childhood musical exposures increase the North Shore University Hospital,
probability of AP in genetically susceptible individuals. Manhasset, New York
The size and complexity of the genetic effect are Nina Kohn
unknown, although a recent analysis [Baharloo et al., Division of Biostatistics,
2000] indicates the presence of major locus effects in North Shore University Hospital,
AP. As in other complex phenotypes, it is likely that Manhasset, New York
combination of genetic mapping approaches will be Elizabeth West Marvin
required to de®nitively demonstrate the role of genetic Eastman School of Music,
factors [Risch, 2000]. Rochester, New York