Harmonizing a tonal melody at the age of 6–15 years

8/12/2019 Harmonizing a tonal melody at the age of 615 years

1/5

ISBN 88-7395-155-4 2006 ICMPC 484

Alma Mater Studiorum University of Bologna, August 22-26 2006

Harmonizing a tonal melody at the age of 615 years

Pirkko PaananenDepartment of Music

University of Jyvskyl40013 Jyvskyl, Finland

[email protected]

ABSTRACTIn this study, 615-year-old participants (N=44)

harmonizations of a tonal melody were analyzed. The

context consisted of two verses of a simple tonal melody in

C major. Four marked keys (C, D, F, G) of the synthesizer,

each of which producing a whole major triad, were used in

improvising an accompaniment as a real-time

performance, recorded with Micro Logic 3.5 / Logic

Education 5.5. Age-related development of harmony and

metre was examined by analyzing the products statistically

(correlation, ANOVA). The cases were clustered withhierarchical cluster analysis.

The frequency of chord C (I) as the final chord, the

distribution of chords to strong metrical positions, and the

number of periodically harmonized measures increased

with age. The number of events and chord transitions

decreased with age, as well as the frequency of the out-of-

key chord D. In addition, seven clusters of varying

developmental levels were found. At the earliest level of

development children focussed on rhythmic impressions of

chords or metre, and coordinated melody and chords

poorly, producing a high number of events. At next level,

coordination of melody and harmony improved, and tonic

closure became common. The chords emerged either at

strong beats, or formed a harmonic riff. In the most

developed products the melody and chords were well-

coordinated, the most frequent chord being C(I), the tonal

closure being common, and the majority of chords

emerging at the strongest beat.

Keywords

Harmony, chords, harmonization, tonality, metre, rhythm,development, children, school age.

INTRODUCTIONFrom the very beginning infants show sensitivity to

melodic contour, familiar scale contexts and sensory

consonance in perception (Trainor, Tsang & Cheung,

2002). However, the development of tonality and tonal

harmony takes several years. Around 5.5 years of age, a

moderate intervallic accuracy, key stability between

phrases is achieved in singing (Davidson, McKernon &

Gardner 1981; Dowling 1988; Kreutzer 2001). Harmonic

perception in childhood has been studied from severalviewpoints: consonance, tonal closure, local chord changes

and implicit harmony. Zenatti (1969) found that 6-year-old

children performed above the level of chance in identifying

a chord that was different between pairs of short

progressions. The ability to identify the chord was

depending on the number of chords in the progression. Hair

(1973) found that first graders were able to perceive

differences between I and V7 chords. OHearn (1984)

found clear improvement between first and third grade in

perceiving a change in harmony in ongoing music. Costa-

Giomi (1994) found that 5-year-old children identified over

50% of the chord changes of a two-chord accompaniment

In: M. Baroni, A. R. Addessi, R. Caterina, M. Costa (2006) Proceedings

of the 9th International Conference on Music Perception & Cognition(ICMPC9), Bologna/Italy, August 22-26 2006.2006 The Society for

Music Perception & Cognition (SMPC) and European Society for theCognitive Sciences of Music (ESCOM). Copyright of the content of an

individual paper is held by the primary (first-named) author of that

paper. All rights reserved. No paper from this proceedings may bereproduced or transmitted in any form or by any means, electronic ormechanical, including photocopying, recording, or by any information

retrieval systems, without permission in writing from the paper'sprimary author. No other part of this proceedings may be reproduced or

transmitted in any form or by any means, electronic or mechanical,

including photocopying, recording, or by any information retrieval

system, without permission in writing from SMPC and ESCOM.


2/5

ICMPC9 Proceedings

ISBN 88-7395-155-4 2006 ICMPC 485

played without melody. In a series of studies Costa-Giomi

(2003) found remarkable improvement in the perception of

implied harmony around 9 years of age under which

performance was poorer in spite of harmonic instruction.

Schellenberg, Bigand, Poulin-Charronat, Garnier & Stevens

(2005) found that 611-year-old childrens judgements ofthe final, target chord in a chord sequence was faster and

more accurate when the target chord was the tonic chord

versus a less stable chord.

While all these studies point to the development of

harmonic perception during the school years, measured

mostly in relatively local contexts, very little is known

about childrens or adolescents abilities to harmonize tonal

melodies. Some case studies, such as Wiggins (1994,

1995), have shown that 1012-year-old children learn to

harmonize and even transpose basic chord transitions to

different keys quite easily after a training period of few

weeks.

AIM OF THE STUDYIn this study, 615-year-old participants (N=44)

harmonizations of a simple tonal melody were analyzed.

Age-related development of harmonic features and chord

rhythm was investigated, as well as the types of

harmonization. The aim was to examine, whether

children/adolescents rely on 1) random or rhythm-based

production of chords, 2) a global sense of key or, 3)

melodic intervals or implicit harmony, as a result of

segmentation and chord recognition (see Jansen & Povel

2004). Furthermore, the role of metre in the development of

harmony was investigated. On the basis of earlier studies of

tonal hierarchy (Lamont & Cross 1994; Paananen 2003) itwas assumed that the youngest (67 yrs) would prefer

random, rhythm based or global key-based strategies. The

ability to segment the melody and select the chord on the

basis of the melodic interval (see Bartlett & Dowling 1980)

or implicit harmonic relations (Costa-Giomi 2003) was not

expected to develop earlier than the age of 89 years. The

studies of tonal closure (Imberty 1969/1981), tonal

hierarchy in perception (Lamont & Cross 1994) and in

melodic improvisation (Paananen 2003), as well as studies

on harmonization (Wiggins 1995, 1996) and harmonic

perception (Costa-Giomi 2003) point to the possibility that

children from the age of 10 onwards would be able to focus

on both melody and harmony in a more differentiatedmanner in perception, and to harmonize a simple tonal

melody.

METHODForty-four children/adolescents took part in the test (N=44;

18 girls and 26 boys). The participants studied at Norssi

School in Jyvskyl, Finland. All, except one 10-year-old

participant, had some experience of playing the synthesizer

or some other keyboard instrument. Five participants had

received at least one year of formal music training, while

the rest had received general music education at school for

12 hours per week in each grade. Six children were in

Grade 0 (=preschool) (range 6.256.50 yrs), five in Grade 1

(range 7.177.83 yrs), six in Grade 2 (range 8.258.92 yrs),

six in Grade 3 (range 9.259.83 yrs), six in Grade 4 (range

10.8010.92 yrs), six in Grade 5 (range 11.1711.83 yrs)

and eight in Grades 89 (range 14.4215.75 yrs).

The melody (fig.1) was a tonal children-song like tune in C

major (oboe sound). However, the tonic is not present until

measure 6, just before the cadence. The most frequently

occurring tones in the melody are G(V), D(II), E(III) and C (I).

The chords including melodic tones are above the melody in

figure 1.

Figure 1.The melody and the chords including melodic tones.

Chord production was reduced to the use of four marked

keys (C, D, F and G) of a synthesizer, each of which

produced a whole major triad (piano sound). With this

method it was possible to harmonize a tonal melody as a

real-time performance and, provide an opportunity to use

more improvisatory or global strategies of production,

with reduced motor demands. The participants explored the

keys and listened to the two verses of the melody before

harmonization. All participants created two harmonizedversions (= two tracks with Micro Logic 3.5 / Logic

Education 5.5.) of the two-versed melody, both of which

were analyzed statistically.

RESULTS

Age-related development of harmonic featuresAge-related development of harmonic and harmonic-

rhythmic features was analyzed by ANOVA (67, 89, 10

11 and 14-15 yrs) and correlation and, the different types of

harmonization with hierarchical cluster analysis. These

features were: the frequency of each chord, the frequency

of out-of-key chord D in the end of measure 5 (-> strong

dissonance), the frequency of the I chord as the endingchord, coordination of melody and harmony in each phrase

at -note level, the number and frequency of chord

transitions and V-I-transitions, number of events, number

of periodically harmonized measures and the frequency of

the eight metrical positions (=the distribution of chords to

metrical positions).


3/5

ICMPC9 Proceedings

ISBN 88-7395-155-4 2006 ICMPC 486

0,15

0,17

0,19

0,21

0,23

0,25

0,27

0,29

0,31

0,33

C D F G

chord

6-7yrs

8-9yrs

10-11yrs

14-15yrs

Figure 2.The harmonic profiles of the four age groups.

Figure 2 presents the harmonic profiles (the frequencies of

each chord) for the four age groups, 67, 89, 1011 and

1415 yrs. A slight preference for chord C is observable

already at 6-7 years of age, however, the use of out-of-key

chord D is almost as frequent as the tonic chord C and, as

frequent as the subdominant chord F. The most developed

profile is the 1415-year-old participants one. The 10-11-

year-olds profile reflects a preference for chord F as the

tonal centre.

Frequency of chord C (I) as the ending chord (Fig 3.)

obtained significant results, correlation with age being

moderate (r= 0.302*; p


4/5

ICMPC9 Proceedings

ISBN 88-7395-155-4 2006 ICMPC 487

Harmonic-metric, developed (mean 146 mos): Tonal centre

C(I), frequency of chord C(I) being 40%, 54% ending with

chord I, coordination of melody and harmony high, uses

more V-I-transitions than any other cluster, number of

events relatively low, 85% of events distributed to metrical

position 1+3+5+7 (1/4-note level) and 68% to metricalposition 1 (strong beat).

Harmonic, early (mean 92 months): Tonal centre C(I),

frequency of chord C(I) being 40%, 58% ending with chord

I, coordination of melody and harmony high, however,

frequency of chord D in the end of measure 5 very high,

number of events high, number of periodically harmonized

measures very low, 39% of events distributed to metrical

position 1+3+5+7 (1/4-note level) but only 19% to metrical

position 1 (strong beat).

DISCUSSIONThe youngest, 67-year-old children, seemed to benefit of

the given method in the sense that it allowed them to focuson more global level of the piece. In general, they focused

rather on the key than local chords and produced more

events than the older participants. As expected, they

produced a high number of events grouped randomly or

into rhythmic patterns, lacking the typical chord rhythm of

tonal music. However, the harmonic early type of

production (mean 92 months) was able to coordinate

melody and chords well, using a strategy of favouring the

tonic chord C and producing a high number of events.

Finding an appropriate chord was easiest in the cadence

already for the 67-year-old children. The oldest age group

(1415 yrs) generally ended the piece with the tonic chord

C. A tendency of placing the chord at metrically strongpositions developed clearly with age, and was common

from age 10 onwards. These older children could also add a

chord in the end of the phrase, or used riff-like patterns.

The frequency of V-I transitions did not increase with age,

which might be due to the given method. It is quite possible

that children needed more time for focusing on local chord

transitions.

The 10-11-year-old children seemed to be puzzled of the

task more than the younger or the older. This age group

produced clearly less events than the younger. They

seemed to have the general idea of what harmony should

sound like, but they did not find the tonal centre C. It ispossible that these children were attending to the melodic

intervals more locally than the younger ones, and were in

the middle of the shift from global to analytic way of

representing tonal harmony.

REFERENCES

Bartlett, J. C. & Dowling, W. J. (1980). The Recognition of

Transposed Melodies: A Key-distance Effect in

Developmental Perspective.Journal of Experimental

Psychology,6, 505-510.

Costa-Giomi, E. (1994). Recognition of chord changes by

4- and 5-year-old American and Argentine children.

Journal of Research in Music Education,42, 68-85.

Costa-Giomi, E. (2003). Young childrens harmonic

perception.Annals of New York Academy of Sciences, 999,477-484.

Davidson, L. & McKernon, P. & Gardner, H. (1981). The

Acquisition of Song: A Developmental Approach. In R. E.

Taylor (Ed.) Documentary Report of Ann Arbor

Symposium. National Symposium on the Applications of

Psychology to the Teaching and Learning of Music.Reston,

Virginia: MENC Reston.

Dowling, W. J. (1988). Tonal Structure And Childrens

Early Learning of Music. In J. A. Sloboda (Ed.),

Generative processes in music. The psychology ofperformance, improvisation and composition (pp. 113-

128). Oxford: Clarendon Press.

Jansen, E. & Povel, D-J. (2004). Perception of arpeggiated

chord progressions.Musicae Scientiae, VIII (1), 7-49.

Hair, H. I. (1973). The effect of training on the harmonic

discrimination of first grade children.Journal of Research

in Music Education, 73, 85-90.

Imberty, M. (1969/1981). Tonal Acculturation And

Perceptual Structuring of Musical Time in Children. In

Basic musical functions and musical ability (pp. 107-130).Stockholm: Publications issued by the Royal Swedish

Academy of Music 32.

Kreutzer, N. J. (2001). Song Acquisition among Rural

Shona-Speaking Zimbabwean Children from Birth to 7

Years.Journal of Research in Music Education, 49 (3),

198-211.

Lamont, A. & Cross, I. (1994). Childrens Cognitive

Representations of Musical Pitch.Music Perception,12(1),

27-55.

OHearn, R. N. (1984). An investigation of the response to

change in music events by children in grades one, three,

and five. Dissertation Abstracts Int. 46: 371A.

Paananen P. (2003).Monta polkua musiikkiin. Tonaalisen

musiikin perusrakenteiden kehittyminen musiikin tuottamis-

ja improvisaatiotehtviss ikvuosina 6-11. (Many Paths to

Music. The development of basic structures of tonal music

in music production and improvisation at the age of 611

years). Dissertation. Jyvskyl Studies in Humanities 10.

University of Jyvskyl, Finland.


5/5

ICMPC9 Proceedings

ISBN 88-7395-155-4 2006 ICMPC 488

Schellenberg, E.G., Bigand, E., Poulin-Charronnat, B.,

Garnier, C. & Stevens, C. (2005). Childrens implicit

knowledge of harmony in Western music.Developmental

Sciente, 8 (6), 551-566.

Trainor, L.J., Tsang, C.D. & Cheung, V.H.W. (2002).

Preference for sensory consonance in 2- and 4-month-old

infants.Music Perception20 (2), 187-194.

Wiggins, J.H. (1994). Childrens Strategies for Solving

Compositional Problems with Peers, Journal of Research in

Music Education,42(3), 232-252.

Wiggins, J.H. (1995). Building Structural Understanding:

Sams Story. The Quarterly Journal of Music Teaching and

Learning,VI (3), 57-73.

Zenatti A. (1974). Perception et appreciation de la

consonance musical par lenfant entre 4 et 10 ans. Sciences

de lArt, 9 (1 & 2), 1-15.

Zenatti A. (1969). Le developpement gntique de la

perception musicale. Monographies Francaises de

PsychologieNo 17.

Harmonizing a tonal melody at the age of 6–15 years

Documents

Transcript of Harmonizing a tonal melody at the age of 6–15 years