Proceedings paper

 

A longitudinal study of the acquisition process of absolute pitch: an effect of subject's age on the process.

Ayako Sakakibara

JSPS Fellowships for Japanese Junior Scientists

3-19-2 Nagasaki

Toshima-ku Tokyo 171-0051 Japan

HQM01603@nifty.ne.jp

 

Introduction

Absolute pitch (AP) is the ability to identify or produce a musical pitch without the use of external reference tones. In this study, some aspects concerning the acquisition processes of AP is investigated.

It has been suggested that everyone initially has the potential to acquire AP. However, almost everyone who exceeds a certain age cannot acquire AP. This "early-learning theory" states that the musical experiences only during a limited early period can be effective to develop AP. According to this theory, we can expect that the learning ability to develop AP is decreased as a person grows older. But little attention has been given to the question about what kind of changes decrease the possibility to acquire AP as one gets older.

I suppose that the acquisition processes of AP would be different by age of the subjects. The purpose of this study is to examine the effect of a subject's age on the acquisition processes of AP.

I trained 6 young children (non-AP possessors) to develop AP by the chord identification training method, and investigated their acquisition processes of AP longitudinaly.

In this training method, children generally start the training at 3 or 4 years old. This age group can be called the "general age". In this study, I trained subjects who started the training when they were younger or older than general age. Subjects of this study were 3 younger subjects starting the training at 2 years old and 3 older subjects starting at 5 or 6 years old.

I would like to clarify the different characteristics of the acquisition processes of AP between the subjects of differing ages. This attempt would offer a key to understanding the changes which decrease the possibility to acquire AP as one grows older.

Method

(1) AP training method.

This study used the chord identification training method, which is the most successful method to acquire AP (Eguchi, 1991). The method consists of tasks for identifying some chords.

Training by the use of chords is considered to be proper to acquire AP. According to the notion that the attributes of tones have two components "tone height" and "tone chroma", the acquisition of AP can be regarded as developing the reference frame of "tone chroma". The use of chords can prompt attention to "tone chroma". In the case of identifying single tone, subjects would tend to pay attention to "tone height". In contrast, the use of chords enables one to make stimulus that "tone height" are similar, but "tone chroma" are quite different. So to identify among chords can make subjects to identify chords depending on their "chroma".

Subjects did chord identification tasks everyday. One session consisted of twenty to thirty trials (it took about three minutes). Subjects had to do four or five sessions per day, thus totaling about one hundred and twenty trials per day. If subjects could identify chords, the number of chords would be increased. At the time when nine kinds of chords are identified perfectly, the AP for every white-key note is acquired(Oura et al,1981).

The training generally took about one year for the acquisition of white-key notes. The analysis of this paper deals with AP of white-key notes (the identification of nine kinds of chords) only. The nine chords were, CEG, CFA, HDG, ACF, DGH, EGC, FAC, HCD and GCE.

(2) Subjects.

In the chord identification training method, children generally start the training at 3 or 4 years old. This can be called as the "general case" (Sakakibara,1998,1999). The subjects of this study were children who started the training at 2 years old (younger cases' subjects) and children who started at 5 or 6 years old (older cases' subjects). 3 subjects of younger cases and 3 subjects of older cases participated this training.

(3) Method of data analysis.

The analysis was done on the data which covered the training records for about one year. The mothers of subjects trained and recorded subjects' responses for every trial. An experimenter instructed mothers of subjects the next training procedure (based on these records), twice a month. A week after the instruction, the experimenter analyzed the records of one whole day.

I analyzed each case longitudinally and illustrated their transitions of cognitive strategies in the acquisition processes. The acquisition processes of younger case group and older case group were compared with the acquisition processes of general case. The aim of this analysis was to explore the differences among subjects of differing ages.

Result

  1. Stages of the acquisition processes of AP.

The acquisition processes of every case could be divided into 4 stages depending on the number of chords of the tasks. The way to divide into stages was same as that of "general case".

The number of chords continued to increase during the training. If subjects could answer every trial of the task correctly, another chord would be added. So the number of chords showed the degree of acquisition. The training process seemed to be differentiated into two types of periods: (a) the period when the number of chords increased and (b) the period when the number of chords didn't increased.

Stage 1 was the period that the number of chords increased rapidly. When the number of chords was six or seven, the chords didn't increase. This period in which chords didn't increase was called Stage 2. After Stage 2, the number of chords began to increase gradually. This period when the number of chords increased gradually was called Stage 3. After Stage 3, the last period when the subjects could answer every chord correctly was called Final Stage.

I illustrated the characteristics of each stage of younger cases and older cases, and compared them to the characteristics of general cases.

(2) Errors of identification tasks.

The training tasks were chords identification tasks. Errors observed in the tasks were classified into the following three categories.

(a) Errors depending on "(tone) height".

The errors were caused by mistaking a chord for a similar one in term of "tone height". They represented the strategy depending on "tone height" (e.g. CEG-CFA).

(b) Errors depending on "(tone) chroma".

The errors were caused by mistaking a chord for another one which component tones were same but different in term of "tone height". In term of "tone chroma", nine chords were differentiated into CEG group, FAC group and GHD group. The mistakes within same group means the strategy depending on "tone chroma" (e.g. CEG-EGC).

(c) Errors indicating no answer .

The subjects had no answer.

What kind of errors occurred in each stage would show characteristics of hearing strategy in each period and their transition.

(3) Investigation of the errors observed in each stage.

[Stage 1]

In the general case, the characteristics of stage 1 were that the percentage of correct answers was very high, and most of errors observed were dependent on "height". Subjects of general case were supposed to use only the strategy to identify chords based on "height".

Stage 1 of the older case group had almost same characteristics as that of the general case, but Stage 1 of the younger case group was different. In Stage 1 of the younger case group, a lot of errors depending on "chroma" were observed. This result showed that younger subjects had a tendency to easily pay attention to "chroma".

[Stage 2]

Stage 2 of every case was the period while chords didn't increase and the percentage of correct answers was low as compared with Stage 1. Stage 2 of the general case was particularly characterized as the period when the subjects began to identify the chords based on "chroma". They were supposed to try to get out of the strategy based only on "height", but they couldn't perfectly use the strategy based on "chroma" yet.

In the case of the younger case group, errors depending on "chroma" already appeared in Stage 1. Stage 2 of the younger case group was the periods when the percentage of errors depending on "chroma" was high.

In contrast, Stage 2 of the older case group had very few errors depending on "chroma". In the older case group, errors depending on "height" were consistently dominant. The result suggested that older subjects tended to identify the chords mainly based on "height".

[Stage 3]

In Stage 3 of the general case, a lot of errors were observed. Especially the percentage of errors depending on "chroma" was high.

The Stage 3 characteristics of the older case group were the same as the general case with respect to low percentage of correct answers. But the details of errors were quite different. Errors observed in the older case group were almost dependent on "height".

In the case of the younger case group, few errors were observed. Younger subjects came to correctly identifying the chords in this stage. Younger subjects always had a few errors depending on "height" throughout the processes.

[Final Stage]

In the Final Stage, there was no difference among cases. Errors of Stage 3 decreased gradually and the percentage of correct answers became 100% in every case.

Figure.1 Percentage of correct answers and three types of errors in Stage 2.

Conclusion

Results showed that errors observed in the training of AP were different among cases differing in subjects' age. The difference means that the strategies children tend to use would change as they grow older.

Younger subjects tended to identify chords based on "chroma". In contrast, older subjects had a strong tendency to identify chords based on "height".

These characteristics suggested that as one grows older, the observed change was a decrease in the tendency to use the "chroma" component. Children would come to depend "height" and not to use "chroma" with increasing age.

If we consider the acquisition of AP as the formation of the reference frame of "chroma", the change suggested by this study is agreement with the principles early-learning theory.

References.

Eguchi,K. 1991 Oto wa rokketto mitaini tondekuru. Tokyo:Niki-syuppan.

Sakakibara,A. 1998 A longitudinal study of the process to acquire absolute pitch. Paper presented at the 5th International Conference on Music Perception and Cognition, Seoul,Korea.

Sakakibara,A. 1999 Zettaionkan syuutoku purosesu ni kansuru jyuudan teki kenkyuu (A longitudinal study of a process for acquiring absolute pitch).Japanese Journal of Educational Psychology, 47,19-27.

Oura,Y.,& Eguchi,K.1981 Is absolute pitch innate or acquired? Paper presented at the colloquium of the International Music Festival, Brno, Czechoslovakia.

 

 Back to index