Proceedings paper


The Relation of body movement and voice production in early childhood music learning

Wilfried Gruhn, Musikhochschule Freiburg, Germany



Observational and experimental research on early music learning (Gordon 1990; Wilson & Roehmann 1990; Deliège & Sloboda 1996; Gruhn 1999) as well as brain studies on music learning (Altenmüller & Gruhn 1997; 1999; Liebert & Gruhn 1999) have demonstrated that the brain holds a powerful potential by its high degree of plasticity which is fundamental to learning. Although all primary sensory cortices are genetically predetermined, the extension and neuronal connectivity within particular brain areas vary over time according to experience and practice. For learning humans profit from the plastic power of the brain which enables it to modulate different functions depending on new demands. The neurobiological potential of the brain makes the "competent infant" (Dornes 1993) whose competence is grounded in the ability to develop dynamic neural networks within a given genetic program and to form mental representations for all kinds of experiences and incoming information.

Since cognitive psychology has described different types of mental representation as "figural" and "formal" (Bamberger 1991) depending upon what aspects the perceptive mind focuses on,

and since EEG studies have identified different activation patterns which can be referred to those types, it is likely to assume that the very nature of infant learning centres in the genetically determined and environmentally stimulated growth of neuronal connections which form genuine musical representations as their neuronal correlates. In a long-term observational study over more than five years (Gruhn 1999; 2000) we exposed groups of young children to various materials derived from Gordon's learning theory (Gordon 1980, 51997) and applied a language acquisition model to the informal teaching and learning process. One of the goals of the longitudinal study was to stress on the interaction of motor skills with musical activities like singing in tune and chanting rhythmically. What educators and musicians like Jaques-Dalcroze, Laban, Jacoby, and Gordon have assumed intuitively or by observation - that different dimensions of body movement reflect musical experience and procedural knowledge - should be proved experimentally.


Preparatory pilot studies

Two pilot studies with three groups of older pre-school children (1. n = 18, M 57 months, SD 6.83; 2. n = 27, M 27 months, SD 6.20) were conducted since 1996. Group 1 (n = 18; m: 9, f: 9; age 4 -5) was taught for more than one year. 1997 two more groups continued with younger children (total n = 27; m: 13, f: 14; age: 2 - 3 years). The main goal of these studies was to explore musical materials appropriate to children's music learning and to develop criteria for data collecting on children's musical growth. Children younger than three years came with their parents or caregivers, and all got an informal instruction once a week for half an hour over a one year period. All sessions were videotape recorded, each child was observed individually, and any developmental change was reported in a protocol. With respect to musical materials the training effect of major/minor based patterns was compared to the effect caused by the combination with other modes (dorian, phrygian etc.). Correspondingly, rhythm patterns in duple and triple meter were compared to those including unusually paired and combined patterns. The broader variety of stimuli produced a slight effect on singing in tune and on discrimination in pitch and duration compared with enhanced training of limited materials, although the data were found to be nonsignificant. With respect to data collection a criterion based observation form (CBOF) was developed containing 45 rating scales referring to five criteria: attention, movement, voice response in tonal context, voice response in rhythm context (imitation, improvisation, creativity), and audiation (see appendix).


Experimental study


Parents of children from birth up to two years from the municipal area of Freiburg volunteered in the study for 15 months from October 1998 until December 1999. Children were selected from a larger sample according to the criteria age (1 year +/- 3 months) and gender (balanced distribution of male and female). From a total of 13 children we lost 4 because of parents' move or other circumstances. Finally, 9 children (M 19.5 months, SD 5.83) completed the study. The social structure was not representative for the average population; all children grew up in a musically active family, and exhibited - as reported by the parents - high musical sensitivity (attracted by music, conscious listening, spontaneous moving along with music). A control group (9 children from a nursery, M 23.2 months, SD 3.15) was matched with the experimental group as to age, gender and social background.


Children with their parents or caregivers participated in an informal instruction once a week for 30 minutes. The entire teaching period of 15 months was segmented into four 10-week-sections. Materials presented were children songs with words, tunes without words, chants with words (nursery rhymes), chants without words, tonal patterns, and rhythm patterns. Songs included all tonalities (major, minor, modes); chants consisted of duple and triple meter and included unusual combinations either way. Whatever type was presented, singing was always accompanied by body movements. However, movement did not necessarily reinforce beats, rather any movement focused on continuous flow and weight. As long as the teacher kept eye contact with a child, (s)he persisted in presenting the same sound pattern to the child. Additional materials (like scarves, balls, hoops, trampoline) were introduced as far as they were apt to support the experience of flow and weight

All sessions were videotape recorded; additionally children were observed individually by two independent judges (interjudge reliability r = .88) using the structured observation form CBOF. Data of any child were averaged for always 4-week periods. For analysis the means of the beginning and ending of each 10-week-section were taken from CBOF. In all, data of 8 measurements (2 for each of the 4 sections) were related to the means of the control group that did not get any musical instruction whatsoever. However, teachers spent approximately the same time with them on the playground and in other classroom activities to gain confidence of the children and to avoid any care effect in the study group. Additionally, parents reported regularly at the end of each section by a questionnaire.



The behaviour over time exhibits slow changes in some areas as type of movement (explorative, imitative, creative), quality of movement (flow, coordination, synchronisation, expressiveness), and voice response (rhythm patterns). (Fig. 1) Since the children started at a very early age, one cannot expect too much progress in voice production (singing and chanting) and little enhancement in movement at the beginning when children begin to stand freely and walk alone. Therefore, only after introducing new models of movement during the first section we find a significant change at the end of that section (flow p = .000; coordination p = .020; synchronisation p = .042). Major and most obvious development of the quality of movement happens only during the second half of informal instruction. In contrast, the readiness for sound reproduction grows slowly. Children need a long time of perceiving structured patterns and processing it before they start imitating them beyond mere babble. Therefore, significant improvement of sound reproduction - if ever observed - is only to be recognised in rhythm patterns.

The development of each particular skill (e.g. coordination, synchronisation, accuracy, consistency, intonation, pitch discrimination, expressiveness) within any criterion (movement, tonal and rhythm response, imitation, improvisation, and audiation) does not proceed continuously, but with accelerations and decelerations. (Fig. 2) Moreover, there is a different progression speed in voice and body movement depending on many biological, social, and environmental influences. From sociology we know about the U-shaped growth which is also reflected by several developmental curves of voice and movement achievement in this study. Any child performs his/her own developmental drift with varying phases of progress and retardation.

Looking for interactions between the various attributes, a Pearson correlation was calculated for each of the 45 criteria. Here, a significant correlation appears only in body movement and voice production at all measurements throughout the four sections. In most cases results are even highly significant (r .80). (Fig. 3) That indicates an important interaction of the way how children use their body for moving and their ability for matching accurate pitches and keeping a consistent tempo. The better they can control their body, the more precisely they can also control their vocal apparatus. With growing experience and practice, the correlation, then, shifts from patterns to songs and tunes at the end of the long-term observation.



Most interestingly, children of the control group show exactly the same direction in their development, but on a lower level although they started from exactly the same level as children in the study group. Significant changes in their musical behaviour over the entire time are nonsignificant except for imitation of movements (p = .009) and coordination (p = .001).

Figure 4

Compared development of movement of children in the study group (S) with those of the control group (C) at the beginning (measurement 1) and at the end of each section (measurements 2 - 5).


Because of their lack of exposure to a musically enriched environment which offers many opportunities to explore their body as well as their voice as a means of expression and communication, the control children could only perform movements according to their biological maturation. In the experimental group that process appeared significantly enhanced.



Although one must take into consideration that correlational findings do not allow causal statements, it is remarkable that the empirical data support a strong interaction between movement and voice production what has already been observed by other music educators. Furthermore, researchers have investigated infants' development of movement and its impact on early childhood music learning (Metz 1989; Blesedell 1991; Hicks 1993; Reynolds 1995). However, it is not easy to adequately interpret the data presented here. One might assume that there is a common neuronal basis for connected mechanisms underlying the performance of movement and vocal sound. But there is no neurological evidence for such an assumption. Rather it is evident that precise control of efferent neuronal transmission that governs motor coordination affects both, general motor skills for moving arms and legs as well as fine motor skills needed for controlling the adjustment of the vocal cords. Movement supposedly facilitates somato-sensori and sensori-motor stimulus transmission and enhances the primary sensori-motor cortex which inversely affects muscular motor processes in the larynx and enables it to react physically to sound by matching a perceived pitch with the vocally produced pitch. If that kind of motor skills is developed properly, correct voice production can function more easily. EEG, MEG and fMRI studies might demonstrate whether there is a neuronal basis for interaction and how it works. As to now one can conclude that music training at an early age is best supported by integrating body movement.

The salient interaction of body movement and voice production can also be interpreted in terms of transfer effects. If we differentiate between internal and external transfer effects (Gruhn 2000), the interaction may be based on an internal effect that connects different brain functions to a more complex network. However, strong research data which support this hypothesis are still missing.

More likely, Condon's observation (1975) that infants immediately after birth respond to mother's voice by synchronous movements parallels our findings. From long-term behavioural observations of newborn he concludes that infants learn their mother tongue even before the age of actual language acquisition by first imitating the structure of movements along with their mothers' speech. Rhythmically structured vocal sequences are - as to Condon - basically perceived as rhythmically structured movement patterns. Investigations on mother infant communications (Malloch 1999/2000; Trevarthen 1999/2000) support the evidence of rhythmically structured patterns in terms of call and response and of childrens' exploration of pitch ranges. Those interactions, which are "built from the units of pulse and quality found in the jointly created gestures of vocalisations and bodily movement" (Malloch 1999/2000, 45), are basic for the development of skills necessary for communication through sound production.


Altenmüller,E. & Gruhn,W. (1997). Music, the brain, and music learning. Chicago: G.I.A. Publ.Inc. (GIML series vol. 2)

Altenmüller, E., Gruhn, W., Parlitz, D. (1999). Was bewirkt musikalisches Lernen in unserem Gehirn? In H.G.Bastian (Ed.). Musik begreifen. Mainz: Schott. pp. 120 - 143.

Bamberger, J. (1991). The mind behind the musical ear. Cambridhe: Harvard Univ.Press.

Blesedell, D.S.(1991). A study of the effects of two types of movement instruction on the rhythm achievment and developmental rhythm aptitude of preschool children. Dissertation Abstracts International, 52 (07), 2452.

Condon, W. (1975). Speech makes babies move. In R.Lewin (Ed.). Child alive. New York: Anchor. pp. 75 - 85.

Deliège, I. & Sloboda, J. (Eds.) (1996). Musical beginnings. Origins and development of musical competence. Oxford: Oxford Univ.Press.

Dornes, M.(1993). Der kompetente Säugling. Frankfurt: Fischer.

Gordon, E.E. (1980). Learning sequences in music. Chicago: G.I.A.Publ.Inc. ( 5th edition 1997)

Gordon, E.E. (1990). A music learning theory for newborn and young children. Chicago: G.I.A. Publ.Inc.

Gruhn, W. (1999). The development of mental representations in early childhood. In Suk Won Yi (Ed.). Music, Mind, and Science. Seoul: Seoul Nat.Univ.Press. pp.434 - 451.

Gruhn, W. (2000). Does brain research support the hope for musical transfer effects? SRPMME Conference, Leicester UK (mscr.)

Hicks, W.K.(1993). An investigation of the initial stages of preparatory audiation. Dissertation Abstracts International, 54 (04), 1277.

Liebert, G., Gruhn, W. et al. (1999). Kurzzeit-Lerneffekte musikalischer Gehörbildung spiegeln sich in kortikalen Aktivierungsmustern wieder. Proceedings of Deutsche Gesellschft für Musikpsychologie. Karlsruhe.

Malloch, S.M. (1999/2000). Mothers and infants and communicative musicality. Musicae Scientiae. Special Issue, 29 - 54.

Metz, E. (1989). Movement as a musical response among preschool children. Journal of Research in Music Education, 37 (1), 48 - 60.

Reynolds,A.M. (1995). An investigation of the movement responses performed by children 18 months to 3 years of age and their caregivers to rhythm chants inn duple and triple meters. Dissertation Abstracts International, 56 (04), 1283.

Tervarthen, C. (1999/2000). Musicality and the intrinsic motive pulse: evidence from human psychobiology and infant communication. Musicae Scientiae. Special Issue, 155 - 211.

Wilson, F.R. & Roehmann, F.L. (Eds.) (1990). Music and child development. St.Louis: MMB Music.


  Back to index