TRANSFER REVISITED:

Re-evaluating the non-musical potential of learning and listening to music

The Copyright of this article is owned by Cambridge University Press. The article first appeared in the British Journal of Music Education, Volume 16 (2), July 1999, pp123-138 and is reproduced here with permission

 

CHAPTERS

  1. Introduction
  2. The Need For Targeting Transfer
  3. Transfer and Automaticity
  4. An Overview of Transfer
  5. Some Recent Reports
  6. Interpreting Results: the need for caution
  7. Uncertain Outcomes: the Kodaly system
  8. Schema Theory and Cognitive Views of Transfer
  9. Conclusion: the need for further research

 
ABSTRACT

The claims of certain reports increasingly appearing in the media suggesting that 'music makes kids smarter' invite critical evaluation. Several other recent studies have been re-examining the hypothesis that musical listening and learning it using methods like Kodaly's may enhance spatial reasoning and social and cognitive growth. In weaving together these interrelated themes, the intention of this article is not only to sensitise music educators to an awareness of transfer's significance and singular character, but to advise discretion in the interpretation of research results. Since the mechanisms of transfer are still not fully understood, future inquiry will have to accommodate many problems facing researchers in the field for whom the formulation of a comprehensive theory of transfer is, as yet, a distant prospect.

 
1. INTRODUCTION

There is nothing new in the claim that musical learning possesses a distinctive ability to transfer its effects to non-musical areas. Throughout history, declarations supporting the beneficial consequences of learning music have appeared and reappeared in various literature and advanced under different guises. They have usually taken the form of highlighting a perceived potential in music for cognitive, emotional, moral, social and general intellectual formation. The essay by Gabor Friss advocating the Kodaly programme of instruction in Sandor's Music Education in Hungary (1975) is illustrative of this tradition:

'The academic record of children attending the music primary schools is much higher than that of children attending the ordinary primary schools, even where normal school subjects are concerned. The reason for this ... lies in music's power to educate, in that it can be used to master other branches of knowledge.' (p 161)

Friss was one of many who held that although the Kodaly educational programme was based on music, its objectives were not exclusively musical:

'... it does not intend to educate towards music alone, but through music serves universal aims.' (p 192)

Statements like these no doubt helped to crystallise, at the time they were written, the long-held belief in the educational and cultural merits obtained from learning music.

Music has long been seen as providing some kind of spur to an individual's general educational and social growth. In classical times philosophical justifications and descriptions of its likely social outcomes can be traced in Plato, whilst nearer our own time, the sociologist Max Weber (1921) has drawn attention to music's social function within the context of the increasing rationalisation of Western economic life under capitalism. Several other references, taken at random and without due regard for chronology, make the point and reflect something of an overall 'mental state' existing over the perception of music's educational and social value. In presenting some kind of counterbalance to alternative symbolic forms, such as mathematics or science, it has been viewed as contributing to the 'intelligence of feeling' (Witkin, 1974) since it is held to possess 'integrative potential' in being able to 'break down barriers' (Paynter and Aston, 1970). Indeed, by way of reinforcing the search for utilitarian pretexts for learning music at school, Paynter included a ten-point list of qualities enumerated by a team from British industry on what industry 'looked for in education' (1982: pp 239-240). The claim here was that music education 'has much to offer industry' since the employers concerned 'could see the importance of music in relation to the qualities mentioned'. These qualities included 'flexibility and adaptability; spatial, mathematical and computational skills; self-awareness and confidence; motor skills such as co-ordination and control; the ability to use initiative in the absence of set procedures' and 'the desire to do a good job.'

Nevertheless, whilst musical learning might defensibly be recognised as offering a contextual catalyst for certain developmental outcomes (Peery and Peery, 1987), the precise relationship between what causes which effect remains a tantalising mystery generating inevitable conjecture. The problematical nature of this relationship demands close scrutiny since it simmers beneath the myriad and often insistent claims periodically advanced within this long-standing cultural tradition.

It could be argued that predicting what will transfer to what may be too much to ask of any theory of transfer; preoccupations with this problem may have misdirected prior research in the field. To be sure, one would like a theory that allowed for fine-grained predictions regarding the 'whats' of transfer. Furthermore, is it the exposure to music itself that supposedly enhances cognitive ability, or rather the musical instruction and learning that may accompany such exposure? More generally, one might ask of such a theory as to whether riding a bicycle would make it easier to drive a car, or if, say, instruction in formal logic would make it easier for a pupil to learn physics. Music educators would like to know if learning to play the oboe, for instance, would make it easier for a pupil to learn how to play the trumpet, or whether singing improves the ability to play a brass instrument. In the absence of such organising frameworks, the complexities of transfer threaten to overwhelm our understanding of its component parts and of the critical factors which often elude the observer. Efforts to frame an over-arching psychological model of transfer would benefit from some kind of analytical taxonomy in the light of which particular skills and configurations of skills could be assayed.

The question of transfer and of its many ramifications, then, is a time-honoured one in psychology and often provokes extreme opinions. Some have claimed it is an elusive rarity (Thorndike, 1913), others that it is ubiquitous in human learning (Hebb, 1949). Gardner (1977) writes of a developmental pattern occurring in the child's increasing capacity to deal with more than one symbol system; as the child attains adolescence there is an increase in the ability to transfer information presented to one modality into a form accessible to others. The monitoring of several sensory channels also increases as information is progressively intertwined from two or more sensory systems.

Many current arguments concerning transfer can be traced back to their roots at the beginning of the 20th century. For example, some people, even today, continue to maintain that the discipline of regular practising of church music at a cathedral choir school somehow spills over into the non-musical areas of a pupil's life. This belief is consonant with a specific viewpoint still held about transfer, namely that the mind acts as a kind of muscle that is susceptible to strengthening by sheer exercise and stretching. Adhered to implicitly for much of the 19th century, the expectation was that an individual's mental faculties could be improved by exertion; this essentially pre-scientific notion suggested a type of 'blanket' transfer of training from one mode to another.

This 'discipline theory of transfer', as it was termed, was, however, to undergo disfavour from those (eg Judd, 1908) holding alternative views about the area. It was attacked by Thorndike who claimed that transfer occurs if, and only if, tasks share 'identical elements' (Thorndike, 1913; Thorndike and Woodworth, 1901). By calling into question the inveterate notion that the classics (the study of Ancient Greek and Latin) provided a cardinal vehicle for 'developing the mind', Thorndike was to proclaim that transfer was narrower in scope than would be predicted by formal discipline and his 'identical elements' theory remained the prominent explanatory concept held about transfer for much of the 20th century. Despite disagreements taking place (Allport, 1937; Orata, 1928, 1945) over what exactly constituted 'identical elements', the position was widely upheld and is still often cited in textbooks as a truism today: if two situations share an underlying deep structure, for example, but differ in their surface manifestations, transfer cannot be expected, whereas if there are surface elements in common, such as physical or perceptual similarity, then transfer will be a 'necessary result'.

 
2. The Need For Targeting Transfer

Whilst the area of the transfer of learning has been accepted as one permeating the wider area of learning itself, it is nevertheless true that an exploration of its multifaceted nature has not received the widespread attention that it deserves in the music education literature. Music educators may wrongly assume that transfer happens somehow by itself, crossing learning domains with ease and extending outwards to general abilities. Furthermore, many studies that do address the topic are embedded in other titles or topic areas.

In fact transfer is far from being an automatic function of either learning or experience, musical or otherwise. A student whose progress is impeded may be unable to generalise or to apply knowledge and skills of whatever kind across a range of situations. If this is so then the music educator is obliged to structure more successful experiences that can promote student independence by concentrating time and attention on the convertibility of knowledge and skills across differing musical contexts. For example, music educators may well be familiar with instances where pupils have been observed executing particular aspects of performance technique in known contexts and with high levels of accuracy, but where subsequent observation of the same pupils has shown them to be confronting major impediments when placed in unfamiliar circumstances. A commonly held belief is that when a pupil has learned to perform a musical figure or passage in one context, the skills and knowledge correspondingly acquired will somehow unintentionally, or even mechanistically, transfer to other situations. Such as assumption is unsound: there exists substantial evidence that it is crucial for educators to target their instruction specifically to facilitate transfer. (Mouly, 1960; Ellis, 1965; Garry and Kingsley, 1970; Postman, 1971; Bower and Hilgard, 1981; Pressley et al, 1987; Brooks and Dansereau, 1987; Brown and Kane, 1988). Transfer depends on the perceived similarity, either surface or structural, between tasks and is highly determined by both the extent and the efficiency of the initial learning. The application of any learned skill is extremely sensitive to the context [1] in which it was learned.

 
3. Transfer and Automaticity

Transfer is more complete the nearer in nature the learning tasks are to each other; indeed, when such tasks radically differ then the investigation into transfer effects can become a hazardous undertaking. Arguably, the researcher into the area would be less likely to encounter formidable difficulties when confining the work to a single learning domain (such as musical learning). More recent inquiry into the precise role of automaticity in performance (eg Neves and Anderson, 1981; Nusbaum and Schwab, 1986) has incorporated much of Thorndike's earlier work and exemplifies how the study and understanding of transfer can be helpful when applied to musical learning and performance.

When learning to play a musical instrument it is crucial for all the sub-components of the activity to become automatic. The existence of these indispensable automatic processes enables one to explain why people are able to play instruments at all, for in order to perform, (at least in the Western tradition), musicians are required to remember clusters of assorted tasks simultaneously: blowing, fingering, decoding the musical text, watching the conductor, following the expressive and dynamic markings, and so on.

Although no invariable view exists over the exact nature of automaticity there is a consensus that the processes involved need minimal or no attention for their execution on the part of the performer. Underwood and Bright, (1996) describe the defining characteristics of automatic activities as ones that develop with extensive practice, are performed evenly and efficiently and are resistant to modification since they remain unaffected by other activities. Future lines of research might well examine more closely the nature of automaticity for there are several ways of considering it. New and unskilled tasks that require conscious control may coexist with familiar, highly practised activities at either ends of a spectrum (Cohen et al, 1990). This view would suggest that although the qualities of automaticity are acquired through practice, attentional factors nevertheless may continue to operate in the learner's cognitive activity. Alternatively, automaticity may involve the hierarchical systemisation of skills with the low-level sub-components automatised through practice (Sloboda, 1985; Smyth et al, 1987). Palmer and van de Sande (1995), in investigating how learning sequences may be planned, have postulated the presence of an interaction between serial influences and the structural boundaries constraining them.

Clearly then the topic of transfer is of paramount importance to music educators, for the area is as diverse and variform as it is for educators in other areas. Discovering how learning can effectively be applied within new settings is a prime task for music educators. But transfer's complex fabric poses additional questions relating to exactly how musical listening might influence an individual's subsequent behaviour and performance success elsewhere. Edwards (1988) sums it up thus:

'Thoughtful performance teachers would admit that transfer pretty well summarises their whole mission in life.' (p 123)

'In the realm of culture the new totalitarianism manifests itself precisely in a harmonising pluralism where the most contradictory works and truths peacefully coexist in indifference' (Marcuse, 1964, p. 61)

 
4. An Overview of Transfer

Transfer of learning means that experience or performance on one task influences performance on some subsequent task. It may take two different forms:

(i) performance on one task may aid performance on a second task (positive transfer), or

(ii) performance on one task may inhibit or disrupt performance on a second task (negative transfer).

Finally, there may be no discernible effect of one task on another, an instance of 'zero transfer', an 'outcome' which can occur either as a consequence of the equal effects of both positive and negative effects cancelling each other, or as a consequence of no effect of one task on the other. (Ellis, 1965)

The most straightforward exemplar would require both an experimental group learning some maiden task (A) before learning a second task (B). The control group does not learn (A) but learns (B). If there is positive transfer from the learning of (A) to (B), then the experimental group should execute task (B) more effectively than the control group. This very basic model does not take account the many variables that would apply, such as age and general intelligence. The assumption has to be made that the two groups were equivalent with respect to certain key variables and that during the interval in which the experimental group learns (A), the control group either does nothing or practises some unrelated task. What this latter group actually does during this period of time is crucial to the interpretation of transfer (Hamilton, 1950; Thune, 1950; Harlow, 1949).

This rather monochromatic account of the phenomenon of transfer offers no clue as to its intricacy, diversity or ambivalence. Nevertheless, given the discussion thus far, it becomes clear that transfer of learning can potentially include the effect of doing anything on the operation of anything else so long as the two actions are not simultaneously performed. Transfer can be defined as the consequence of learning skills, knowledge or attitudes on later learning of other skills, knowledge or attitudes. Questions relating to the manner by which such skills, knowledge or attitudes can be applied in new settings are central ones for the investigator into transfer effects.

As stated earlier, Thorndike and Woodworth (1901) tried to unravel some of the controversial problems of transfer and to systematise explicit theories of it.

Then for some time the field of enquiry became embedded in more general cognitive theories that dealt with transfer indirectly by implication only. For example, from an information processing perspective, one could deduce that transfer depends on how a memory search is initiated, the kinds of memory modes accessed, and the extent of their inherent connectedness to other memory modes. From the perspective of schema theory, referred to later in this paper, it could be argued that transfer depends on the activation of relevant anticipatory schemata. More recently, issues relating to the area have again surfaced and become explicit in several contexts, including such matters as the cultivation of intelligence, the training of strategies for learning (Nickerson et al, 1985; Weinstein and Mayer, 1985), the study of how programming can affect cognitive skills and the study of how inert knowledge could be effectively utilised in new settings (Pea and Kurland, 1984; Bransford et al, 1986). These very different contexts call for a broad perspective of a reality which can so easily escape recognition below the surface of many psychological and educational investigations. For transfer is far from being a unitary phenomenon, and there is still no agreement on exactly how it occurs. Indeed it may occur by different routes dependent on different mechanisms, or networks of mechanisms: for example, one mechanism may be more appropriate for the transfer of explicit knowledge and strategies, whereas another may be more so for more general activities.

 
5. Some Recent Reports

The issue of transfer has become a topical one. Several reports have appeared in the press testifying to alleged benefits to young children when they are exposed to music. The sincerity and conviction of claimants is evident:

'I found that through musical activities children may develop many of the skills required for literacy and that children who attended pre-school groups fared better in reading tests than those who didn't.' (A teacher of eighteen years experience who has devised what is described as 'a pioneering literacy programme' through music. Report in a regional newspaper. 22 September 1998)

'Scientific research in the 20th century has revealed that music plays a key role in the functioning of the brain and behavioural psychologists have shown how it can aid the learning process.'

(A report in the Times Educational Supplement, 24 April 1998)

'Researchers at California University concluded that music modifies circuits in the brain leading to improved thinking skills ... primary pupils in Rhode Island, USA, made better progress in reading and maths as a result of following an enhanced arts curriculum with music teaching.'

(A report in the Times Educational Supplement, 1 May 1998)

It has been alleged that musical learning 'improves the cerebral software' (Rauscher et al, 1993), that 'learning arts skills forces mental stretching useful to other learning areas such as maths' (Gardiner et al, 1996) and we are told that 'the established findings of educationists show that music in schools has measurable effects on learning' (Everitt, 1998). A report in the Guardian (25 May 1996) claimed that 'listening to Mozart improved spatial reasoning skills' an hypothesis that was investigated in the UK when 11,000 children in 250 schools were randomly split into three groups, one listening to a Mozart quintet, one to pop music, and one to live conversation from the radio. The article ends by saying 'in tests of spatial reasoning, the pop listeners scored 56% compared with 52% for the others.' It also included the statement that 'boys in cathedral schools performed better in all subjects, even when they were not academic, because they were learning music'. The claims of Rauscher, Shaw and Ky working at Irvine, California, already alluded to, are also echoed and described in this report. Students in the Rauscher experiment had been given three sets of standard IQ spatial reasoning tasks with each task prefaced by ten minutes of:

  1. listening to the two-piano Sonata, K.448 of Mozart

  2. listening to a relaxation tape; or

  3. silence

The report asserts that the performance of the students on the spatial reasoning tasks that followed the Mozart listening was better than for the other two listening conditions, whilst also declaring that the enhancing effect of the music was temporary and therefore 'unlikely to exist much beyond the ten to fifteen minutes during which the students were performing the spatial task'. As it had originally appeared in Nature (14 October 1998) the report had ended with the authors hoping that since only one musical example had been used, various other styles and composers should also be investigated.

Two other reports can be cited here. Firstly, the Rhode Island report by Gardiner et al (1996). This described how work that had been introduced to groups of first-grade children, aged between five and seven years, to music and a visual-arts curriculum, had been co-ordinated with other subject areas with the emphasis on sequenced skill development. Many of the 'test' group children had not performed well at kindergarten as compared with the 'control' group, the latter exposed to what the researchers describe as 'just the usual visual arts and music training typically encountered in the public-school system of the States'. Apparently, the 'test' group had caught up with the 'control' group after a mere seven months, even outstripping the latter in mathematics, despite the fact that the 'test' group had contained children of whom less than 40% (as against the 'control' group's 70%) had scored marks at, or above, the national average in post-kindergarten testing. The credit for this distinction was accorded to the Kodaly method that had been used for the music teaching in the school.

Secondly, a Swiss report (Spychiger et al, 1995) described an interregional school experiment and was entitled, 'Better Formation with More Music'. In this experiment fifty classes had received five weekly music lessons in place of the usual two. At least three of the additional music lessons had resulted in a reduction in the number of lessons of mathematics or mother-tongue instruction, the hypothesis being that the increased music teaching would over-compensate for the diminution of weekly hours instruction in other subjects, ie that in spite of fewer lessons, the pupils would learn as much as other students, and gain in musical competence as well as in concentration, creativity, motivation, even perhaps in physical development. In suggesting that the report title, 'Does more music teaching lead to a better social climate?' warranted an affirmative answer, the authors professed to detect an increase in sociographic integration[2], concluding that the increase in music education could have enhanced the effect of social coherence even though their observations did not include detecting any traceable enrichment in feelings of social community as a whole. 'Strong statements could not be the aim of this paper', the authors admit, indicating their view that the derivatives of music education are rather more complex than is usually assumed; increased music teaching needs time to develop its impact on such specifically non-musical factors as social behaviour.

Notwithstanding the reservations felt by these authors, the general thrust remains clear; music does help to improve task performance in other areas, and even if it does not, then it certainly does not hinder such performance. This viewpoint embraces one made earlier by Hanshumaker:

'School time spent on music and other arts activities has no negative effect on academic achievement as reflected by standardised tests.' (1980)

Undoubtedly these reports and others like them, as well as their accompanying claims, have had an impact on public opinion in the States and elsewhere. This change of heart, if such it is, has been tellingly encapsulated in a picture of President Clinton presenting one of his saxophones to a youngster in April 1998 to signify the inauguration of a 'Save the Music' campaign pinpointing the necessity for reinstating music tuition in the high schools of the United States.

 
6. Interpreting Results: the need for caution

Rauscher claims the existence of a correlation between music and the operation of the brain's higher functions, such as reasoning and recognition. By listening to Mozart, or by playing a keyboard before being tested, students will perform better on spatial-temporal tasks since, so the argument goes, music directly affects cognitive processing by priming neural circuits situated near to those used for aspects of cognitive processing. However, as Hallam and Katsarou have stated (1998), Rauscher's interpretation refers to only a very restricted aspect of cognitive functioning since the effects appear not to extend to other aspects of logical reasoning. Moreover, it is unclear why a single piece was used in the testing and why the assumption is tacitly made that 'complex' music (which is alleged to enhance abstract reasoning) is held to be synonymous with 'classical' music, and that 'music lacking complexity' (alleged to interfere with abstract reasoning) is synonymous with 'non-classical' music. Furthermore, the relationship between the tests is never explained, nor is the unsubstantiated statement made about the way music might actually be assimilated:

'... as musicians may process in a different way from non-musicians, it would be interesting to compare the two groups.' (Rauscher et al, 1993.)

Possibly an implied reference is being made here to work by Pollard-Gott who demonstrated (1983) how repeated listening to music examples can result in thematic extraction by professional musicians, ie those possessing a musical background (in the conventional sense of that term) appeared more readily to perceive the significance of a composition's thematic material than musically naive subjects. Clarke and Krumhansl (1990) and Deliege and El Ahmadi (1990) have also investigated this area, the work of the former revealing how musically trained subjects tend to segment music, identifying the significance of boundaries existing within a piece of music, whilst that of the latter has focussed on the presence of two organising principles at work in the way listeners perceive musical form, namely those of sameness and difference. Significantly perhaps, both musicians and non-musicians revealed basic congruities in their respective perceptions of music, a finding which calls into question Rauscher's unqualified supposition suggesting the contrary. Rauscher's work seems to assume such 'differences' when, in fact, the evidence for them is not compelling. Whilst it may have been apparent to his team of researchers that listening to Mozart (or being involved in playing a keyboard) immediately prior to testing gave an incisive edge to the performance of spatial-temporal tasks requiring reasoning, the test itself was too wide-ranging and lacked the sufficient focus to lend it entire credibility; future work will have to take other variables into account. Such variables could include the kind of music used, the personality, temperament and recent life-history of the individual listener, the listening environment and context, metacognitive factors, and the relationship between observable short-term gains and any symptomatic long-term effects - indeed the very robustness of transfer itself as a phenomenon.

Underlying weaknesses in research design as well as ill-considered verdicts over music's potential for activating the 'cerebral hardware' have left unexplained central questions relating to the nature and existence of these connections. There is still uncertainty as to how music listening skills might transfer to listening situations beyond the music classroom. The Rhode Island report is undermined by the fact that the pupils did not start the experiment at the same age. Moreover, it avoids any mention of the socio-economic dimension. The Swiss experiment, though intriguing, makes unscientific judgements and fails to demonstrate conclusively the consequences of increased music education on several dimensions of the social climate; the report contains too many 'might's' and 'possibly's' and other such conditionals in the language to make its evidence appear compelling. As the authors themselves are forced to acknowledge, the likely derivatives of music learning seem relatively cursory when compared with the high expectations that common-sense might lead one to harbour. But to them the results of this somewhat unrigorous three-year experiment were clear:

'The results indicate that the benefits of EMT (Extended Music Teaching) ... indicate the development of an interest in music. No negative or scholastic effects were observed amongst the entire experimental group in spite of the reduced number of lessons in certain subjects.' (Zulauf, 1993)

Accordingly, many of the studies gaining national press attention fall short by not convincingly addressing the question as to whether it is the musical learning at all that has caused pupils to perform differently in other areas. Whilst visibly plausible in appearing to pioneer breakthroughs in our grasp of these areas, they have, in fact, presented a case not borne out by the data assembled. Moreover, merely establishing certain relationships between learning domains and 'abilities' tells one little about any causal linkages that might, on the balance of probabilities, exist between them. Of far more interest is the question as to why such reports are thought to be necessary in the first place.

One would reasonably expect musicians, music teachers and the music industry to extol the educational advantages of learning and playing music. Proponents of some of these studies are often well-intentioned music educators engaged in a campaign to reassert music's threatened position within the curriculum. At a time when coercive economisation endangers the very existence of many music programmes (as true in the UK as it is in the United States), this compulsive willingness to embrace any such research is certainly understandable. But in the worst cases of this ever-growing line of enquiry, we can see instances of methodologically flawed enquiry misleadingly upheld as 'truth'. Therefore caution and discretion are both necessary in interpreting these reports, for sometimes in such work, as in much else in life, we tend to see only what we are looking for.

 
7. Uncertain Outcomes: the Kodaly system

I began this article with a reference to the Kodaly system. Long suspected to have important implications for learning effectiveness in non-musical domains (eg Kokas, 1969; Hurwitz et al, 1975), this form of musical education offers pupils a multifaceted perceptual and cognitive training and is aimed at instructing all children, not just those perceived as being musically gifted. Besides using a variety of techniques including hand-signs, clapping, games as well as reading staff notation, the Kodaly system places a strong emphasis on choral activities since it is through singing, it is claimed, that pupils receive their most meaningful experiences. Furthermore, the development of rhythmical skills is emphasised in the programme with the teaching of folk songs from which the instructor abstracts basic rhythmic as well as melodic units. Through a step-by-step process, the Kodaly instructor increases the child's awareness of these rhythmic entities and in turn uses them to create new rhythmical constructions which the child may use in new combinations. The progressive development of singing and other musical skills thought to be necessary for the promotion of pupil competence is reinforced by what its adherents see as a logical ordering of subject content accentuating continuity and sequencing. Musical activities are mostly teacher-directed and didactic, whilst the materials evolve through composed music for children to the 'high art' music of such composers as Mozart, Beethoven and Schubert.

If any mode of teaching and learning music could be said to unlock the door leading to enhanced pupil non-musical learning, conventional wisdom might suggest that this system held the key to it. The kind of systematic rhythmic, melodic and choral development provided by the training must surely be able to influence the child's sequencing ability in such a way as to improve other areas of cognitive development. And initially, it was the case that much of the exploratory research, particularly in Hungary where the programme began, professed to show that the Kodaly system was exerting a dynamic effect on such areas as visual observation, spelling, language learning and movement. These conclusions tied in with other studies reporting correlations between musical ability and other facets of intelligence, such as creativity (eg Barkoczi and Pleh, 1982). But again it has to be said that it is one thing to establish these 'relationships' and quite another to identify causal links between them. (Herbert, 1974; Wolff et al, 1975; Douglas, 1976; Kalmer and Benis, 1979; Karma, 1982; Laczo, 1985; Cropley, 1991). The consequences of using the Kodaly programme are mixed. Bain (1978) reported a positive transfer to body percept through the use of a modified Suzuki and Kodaly instructional programme, and Hurwitz et al (1975) claimed that Kodaly instruction positively affected temporal/spatial abilities but neither of these studies reported any transfer to verbal abilities. Bain believed that early music training could increase a child's sensitivity to his/her 'inner dynamics' and create a more flexible means to achieve cognitive goals. However, he observed that 'the heuristic value of specific educational experiences remained limited to one domain' (p 80). Positive effects of such instruction on verbal ability have indeed been demonstrated, but with special populations of disadvantaged or neurologically impaired children (Aten el al, 1984; Pirtle and Seaton, 1973). Both of these latter studies used music learning tasks specifically targeted at verbal enrichment. But neither Lauder (1976) nor Sharman (1981) could show transfer effects from music activities to reading skills even though the music instruction had been intentionally interlocked with the reading instruction. In the light of such uncertain outcomes, as Wolff has stated (1978), it would seem that whilst there may be some observable, quantifiable effects of music education on the development of understanding and cognitive abilities, the argumentation used in so much of the work to support the claim appears inexact and the conclusions drawn largely improbable. For no general theoretical approach as yet exists that would adequately account for the assumed non-musical effects of music education. Whether it is imparted through the Kodaly system, or not, music instruction per se cannot be guaranteed to fabricate specifically desired and discernible non-musical by-products even when that instruction has complied with the targeting imperative and has been highly structured towards that objective.

 
8. Schema Theory and Cognitive Views of Transfer

Being able to discriminate between assorted strands of aural information is a primary requisite for successful musical learning. Musical perception is analogous to visual perception in that it constitutes a process of intelligent reconstruction with learning taking place when pupils gain greater freedom in being able to elaborate their representations in the discovery of new relations presented to them within sound stimuli. The human perceiver processes information in many ways depending on which features are given priority to, or, at least, are accessible to the learner. The prime focus for such an engagement with sound material is affected by the defining contexts in which a particular figure is embedded: a new context would suggest new priorities and reveal new features.

At the heart of modern cognitive psychology lies the concern with the flow of information, namely its acquisition, transmission, storage, retrieval and transformation in the learner's mind. In this conception of learning, memory is seen as a highly structured and interconnected system with understanding occurring when new information is related to existing knowledge; transfer is therefore likely to occur when the learner has encountered the relevant prior information. (Royer, 1979; Shuell, 1986) Schema theory, which has grown from this area of psychology, is more accurately a wide range of theories addressing the nature of this structure. Schemata are connected and hierarchically managed frameworks constituted of other (sub) schemata like data, abstractions, percepts, concepts, mental images, 'know-how' and other assorted variables (Howard, 1987) and occupy an uppermost position in learning. In being both abstractions from experience as well as representations of it, they facilitate pattern recognition and at the broadest level affect the amount of information an individual actually takes in.

Singley and Anderson (1989) stated that memory consists of two principal categories: declarative and procedural. Declarative knowledge requires 'proceduralisation' by being executed directly without the student having to think about it. An example may help to clarify this further.

Music's inherent curriculum status can become downgraded in the eyes of those that would hold a view of it that it is somehow 'weak' in 'hard facts'. This viewpoint underestimates the weight attached to the very knowledge that a person needs to be creative in the first place. In music, to be 'creative' one has to know certain aspects of knowledge - a crude viewpoint would hold that such aspects would include familiarity with the theory and conventions of staves, clefs and keys; in whatever style the music is created there will always be supporting codes comprising 'inert' knowledge that has to be known. Reading music demands that the learner understands the nature of written material, and is, itself, a domain-specific part of the more general ability to read.

All knowledge is constructed hierarchically (Neves and Anderson, 1981) and initially encoded as declarative knowledge, ie as sets of facts. An individual beginning to learn a musical instrument has to know what to do, where to place the hands, how to blow or to hold the instrument. But this essentially declarative knowledge, whilst indispensable to the learner is, by itself, insufficient to maintain a performance since it requires transforming into procedural knowledge. Any performance solely based on declarative knowledge will be uncertain and weak for the student will be having to think, for example, where to place the hands when playing a keyboard instrument. By proceduralising declarative knowledge the task's memory load is reduced and the student can then attend to the aesthetic, interpretative qualities of the performance; the task becomes less a matter of what to do than how to do it. Whilst declarative knowledge is concerned with facts and things, procedural knowledge is that leading to action, a fundamental distinction having implications for the teaching of intellectual skills and productions. In the sphere of music, procedural memory may take the form of intellectual procedures such as motor schemata or the analysis of harmony. (Schmidt, 1975; LaBerge, 1981) And just as the acquisition of declarative knowledge will not ensure that procedural knowledge has been obtained, so the reverse is also true. Effective teaching means that the teacher has recognised something of the intrinsic mutuality and interplay of these two important provinces. (Singley and Anderson, 1989; Postman, 1973; Milner, 1962)

Schema theory has been applied to musical learning in the work of both Jones (1982) and Bharucha (1987); their ideas are intriguing since it is likely that they could hold strong implications for music educators for an understanding of transfer. Jones has advanced a concept of 'ideal prototypes', ie long-standing conceptual models moulded from autobiographical experiences with music. All of us have a 'reservoire' of aural experiences, banks of data constructed from a life-time's involvement of varying degrees with an extensive terrain of listening, performing, and, perhaps, composing transactions. Jones sees his prototypes as forming 'expectancy schemata' serving as templates for comparison with musical input. For example, I may have formed an ideal prototype of music derived from a particular style or historical period. Whether or not the music is known to me, my processing of it would allow me to correlate any incoming musical material with any expectancies I had about it through the activation of expectancy schemata. The links between this view of prototypes and the area of transfer of learning are considerable for Jones' work goes some way towards explaining why one can classify musical styles, periods and performances. A strong affinity with Hans Keller's notion (1970) of foregrounding and backgrounding in music is also suggested:

'The background of a composition is both the sum total of the expectations a composer raises in the course of a piece without fulfilling them and the sum total of those unborn fulfilments. The foreground is simply what he does instead.' (Keller, 1970)

The concept of expectation is woven into the writings of musical theorists and is a salient aspect of the temporal dynamism of music (Meyer, 1956; Schoenberg, 1954); 'a dominant chord, for example, indicates that the tonic has yet to come' (Schenker, 1906). Any musical context establishes a graded array of expectations that offer the basis for ambiguity and for varying degrees of resolution. As described by Meyer, expectations may, or may not, be conscious, but their generation and subtle violation is crucial both to the psychology of emotion and to music aesthetics (Mandler, 1984).

Bharucha's work looks at computer simulation models of spreading activation of schemata and the effects of priming listeners with tonal contexts on their musical processing. He acknowledges that, as listeners, we are broadly familiar with Western patterns of expectation (there are powerful and authoritative accounts explaining them in terms of chord and key relationships) but goes further by investigating the question as to how these expectancies might be generated and what the psychological operations could be that elicit them. Bharucha views schematic representations as components of a larger, interconnected system that includes sequential memory traces of specific pieces of music. The acquisition of schemas mostly occurs in a cultural context; those that are dependent on perception and cognition are valid only for those individuals who have acquired these schemas through continuing contact with the style of music specific to their culture. Accordingly, Bharucha's work is restricted to a proposed harmonic model that characterises the internalised structural apparatus of an average Western listener who has had a life-time's exposure to habitual chordal relationships. However, he writes:

'No compositional prescription or value about tonal music is in any way entailed. The claim is only that if musical context engages a schematic representation, perceptual facilitation will occur automatically in accord with the representation's structural constraints ... music that at least mirrors an internalised representation is likely to be more accessible to the average Western listener.' (Bharucha, 1987, p 27)

 
9. Conclusion

The overall thrust of this article has been to suggest that:

(i) the formulation of an overarching theory of transfer remains a distant vision

(ii) the results of research on general transfer of musical learning have so far shown inconsistency; and

(iii) the claims being made by some people about the effects of learning music on non-musical areas should be treated with circumspection.

The degree to which music can be said to affect cognitive processing at all requires continued and sustained inquiry from researchers. Gifford (1988) has stated that it is unlikely that an individual's mental faculties are improved through what he calls 'the power of music'. Instead, he strongly argues that a sense of achievement in music performance could induce positive attitudes whereby both concentration and memory may develop through being transferred to other curriculum areas. This perspective has also been voiced by Gagne who held (1978) that good study habits acquired in any domain of learning possessed the potential for transferring to other domains.

However, the cognitive argument cannot be entirely discounted. As has been suggested (Hallam and Katsarou, 1998) music's potential to influence mood may be accomplished through relatively primitive brain mechanisms which, while they can be brought under some conscious cognitive control, may well operate independently. Mood might therefore be altered without the influence of interfering memories disturbing cognitive processing. Calming, relaxing music appears to exert a positive effect on children when they are remembering words from sentences, whilst music that has been thought of as aggressive exerts a negative influence on cognition in relation to recall. How these effects occur remains an open question. Future work could impact on our comprehension of cognition; disentangling transfer's intricate and multiple mechanisms and probing into the nature of musical processing are long-term tasks linked to existing knowledge about memory, reasoning, categorisation and problem-solving. Greater attention is needed in research work over the issue of the very potency of transfer's dynamic; given that a task has been mastered, for how long will transfer persist? In memory tasks delay sometimes even facilitates performance (Thorndyke and Hayes-Roth, 1979). Future examination of transfer should also differentiate between its many transitory effects and those consequences that are more long-term. Furthermore, if the key to an understanding of transfer's behaviour in musical listening tasks lies at the cognitive level, which seems likely, then some of its worthwhile corollaries may be accomplished by the priming of neural circuitry, while others may be mediated more by mood and therapeutic arousal. Although much in these fields of enquiry remains speculative there is every reason to hope that persistent dissection of transfer's mysterious textures will encourage music educators to wrestle with these eminently practical matters.

 
End Notes

[1] The context of a transfer situation will be determined by events occurring close to it in time or that are retrieved from memory during the performance of the transfer task. The context of two situations can affect perceived similarity. Any salient shared component, whether surface or structural, will increase the likelihood that two situations will be viewed as being similar by the learner. To some extent the learner's expertise in the area will determine whether the salient similarities noticed are surface or structural.

[2] A sociogram is a diagram in which group interactions are analysed on the basis of mutual attractions or antipathies between group members.

 
BIBLIOGRAPHICAL REFERENCES, SUPPORTING MATERIAL and RELATED READING

AEILLO, R. and SLOBODA, J. (1994) Musical Perceptions. Oxford University Press.

ALLPORT, G. (1937) Personality: a psychological interpretation. NY Holt and Company.

ATEN, J., SMITH, G. and TUNKS, T. (1984) Music in a remedial oral language programme: final evaluation of cycle I. Paper presented at the Texas Music Educators National Conference. San Antonio.

BAIN, B. (1978) The cognitive flexibility claim in the bilingual and music education research traditions. Journal of Research in Music Education. Volume 25 (2) pp 76-81.

BARKoCZI, I. and PLeH, C. (1982) Music Makes a Difference: the effect of Kodaly's music training on the psychological development of elementary school children. Kodaly Pedagogical Institute of Music, 21473 Petofi Nyomda, Kecskemet, Hungary. Translated by Steiner Szilveszterne and Csaba Pleh.

BHARUCHA, J. H. (1987) Music cognition and perceptual facilitation: a connectionist framework. Music Perception. Volume 5 (1) pp 1-30.

BOWER, G. H. and HILGARD, E. R. (1981) Theories of Learning. (Original edition, 1948). Prentice Hall.

BRANSFORD, J. D., SHERWOOD, R., VYE, N. and REISER, J. (1986) Teaching thinking and problem solving. American Psychologist. Volume 41 pp 1078-1089.

BROOKS, L. W. and DANSEREAU, D. F. (1987) Transfer of Information: an instructional perspective. In Cormier, S. M. and Hagman, J. D. (Eds) Transfer of Learning: contemporary research and applications. Academic Press. Harcourt, Brace, Jovanovich. pp 121-149.

BROWN, A. L. and KANE, M. J. (1988) Preschool children can learn to transfer: learning to learn and learning from example. Cognitive Psychology. Volume 20 pp 493-523.

CLARKE, E. and KRUMHANSL, C. (1990) Perceiving musical time. Music Perception. Volume 7 pp 213-251.

COHEN, J. D., DUNBAR, K. and McCLELLAND, J. L. (1990) On the control of automatic processes: a parallel distributed processing account of the Stroop effect. Psychological Review. Volume 97 pp 332-361.

CROPLEY, A. J. (1991) Improving intelligence by fostering creativity in everyday settings. In Rowe, H. A. H. (Ed). Intelligence: Reconceptualisation and Measurement. Hilldale NJ.

DELIeGE, I. and EL AHMADI, A. (1990) Mechanisms of cue extraction in musical groupings: a study of perception of Sequenza IV for solo viola by Luciano Berio. Psychology of Music. Volume 18 (1) pp 18-44.

DOUGLAS, P. (1976) An investigation of the relationship between musicality and intelligence. Psychology of Music. Volume 4 (2) pp 16-31.

EDWARDS, R. H. (1988) Transfer and performance instruction. In Fowler C. (Ed). The Crane Symposium: toward an understanding of the teaching and learning of music performance. NY Potsdam College, State University of New York.

ELLIS, H. C. (1965) The Transfer of Learning. MacMillan.

EVERITT, A. (1998) 'Cerebral software'. Opinion. Times Educational Supplement. 24 April 1998 p15.

FRISS, G. (1975) The Music Primary School. In Sandor, F (Ed). Music Education in Hungary. Boosey and Hawkes. pp 145-192.

GAGNe, R. (1978) The Conditions of Learning. 2nd Edition Holt, Reinhart and Winston.

GARDINER, M. F., FOX, A., KNOWLES, F. and JEFFREY, D. (1996) 'Learning improved by arts training'. Scientific correspondence. Nature. Volume 381 23 May 1996 p 284.

GARDNER, H. (1977) Senses, Symbols, Operations: an organisation of artistry. In Perkins, D. and Leondar, B. (Eds). The Arts and Cognition. John Hopkins University Press. p 94.

GARRY, R. and KINGSLEY, H. L. (1970) The Nature and Conditions of Learning. (Original edition, 1946) Prentice-Hall.

GIFFORD, E. F. (1988) An Australian rationale for music education revisited: a discussion on the role of music in the curriculum. British Journal of Music Education. Volume 5 (2) pp 115-140.

HALLAM, S. and KATSAROU, G. (1998) The effects of listening to background music on children's altruistic behaviour and success in memorising text. Paper presented to the British Educational Research Association, Belfast, April 1998.

HAMILTON, C. E. (1950) The relationship between length of interval separating two learning tasks and the performance of the second task. Journal of Experimental Psychology. Volume 40 pp 613-621.

HANSHUMAKER, J. (1980) The effects of art education on intellectual and social development: a review of selected research. Bulletin of the Council for Research in Music Education. Volume 61 pp 10-27.

HARLOW, H. F. (1949) The formulation of learning sets. Psychological Review. Volume 56 pp 51-65.

HEBB, D. O. (1949) The Organisation of Behaviour. NY Wiley.

HERBERT, G. F. (1974) An initial assessment of the performance of a group of 150 Grade One children whose programme has included Kodaly-based music training. Slow Learning Child. Volume 21 (1) pp 15-23.

HOWARD, R. W. (1987) Concepts and Schemata: an introduction. Cassell.

HURWITZ, I., WOLFF, P. H., BORTNICK, B. D. and KOKAS, K. (1975) Non-musical effects of the Kodaly Music curriculum in primary grade children. Journal of Learning Disabilities. Volume 8 (3) pp 45-51.

JONES, M. R. (1982) Music as a stimulus for psychological motion: II An expectancy model. Psychomusicology. Volume 2 (1) pp 1-13.

JUDD, C. H. (1908) The relation of special training and general intelligence. Educational Review. Volume 36 pp 28-42.

KALMER, M. and BENIS, M. (1979) The effect of musical training on the development of qualification concepts in nursery schoolchildren. International Kodaly Society.

KARMA, K. (1982) Musical, spatial and verbal abilities: a progress report. Journal of Research in Music Education. Volume 9 (3) pp 102-114.

KELLER, H. (1970) Towards a theory of music. The Listener. 11 June 1970.

KOKAS, K. (1969) Psychological tests in connection with music education in Hungary. Journal of Research in Music Education. Volume 9 (3) pp 102-114.

LaBERGE, D. (1981) Perceptual and motor schema in the performance of music. Documentary Report of the Ann Arbor Symposium. Reston: Music Educators National Conference.

LACZO, Z. (1985) The non-musical outcomes of music education: influence on intelligence? Bulletin of the Council for Research in Music Education. Volume 85 pp 109-118.

LAUDER, D. C. (1976) An experimental study of the effects of music activities upon reading achievement of first grade students. Unpublished doctoral dissertation. University of South Carolina, Columbia.

MANDLER, G. (1984) Mind and Body: psychology of emotion and stress. NY Norton.

MEYER, L. (1956) Emotion and Meaning in Music. University of Chicago Press.

MILNER, B. (1962) Les troubles de la memoire accompagnant des lesions hippocampique bilaterale. In Physiologie de l'hippocampe. Paris: Centre National de la Recherche Scientifique.

MOULY, G. J. (1960) Psychology for Effective Teaching. (3rd ed, 1973) Holt, Rinehart and Winston.

NEVES, D. M. and ANDERSON, J. R. (1981) Knowledge compilation: mechanisms for the automisation of cognitive skills. In Anderson, J. R. (Ed) Cognitive Skills and their acquisition. Hillsdale NJ. Erlbaum.

NICKERSON, R., PERKINS, D. N. and SMITH, E. (1985) The Teaching of Thinking. Hillsdale NJ Lawrence Erlbaum Association Incorporated.

NUSBAUM, H. C. and SCHWAB, E. C. (1986) The role of attention and active processing in speech perception. In Schwab, E. C. and Nusbaum H. C. (Eds). Pattern Recognition by Humans and Machines. NY Academic.

ORATA, P. T. (1928) The Theory of Identical Elements. Columbus. Ohio State University Press.

ORATA, P. T. (1945) Transfer of training and educational pseudo-science. The Mathematics Teacher. Volume 28 pp 265-289.

PALMER, C. and van de SANDE, C. (1995) Range of planning in music performance. Journal of Experimental Psychology. Volume 25 (5) pp 947-962.

PATER, W. (1910) The Renaissance. MacMillan. pp 134-135.

PAYNTER, J. and ASTON, P. (1970) Sound and Silence: classroom projects in creative music. Cambridge.

PAYNTER, J. (1982) Music of the Secondary School Curriculum: trends and developments in class music teaching. Cambridge. pp 239-240.

PEA, R. D. and KURLAND, D. M. (1984) On the cognitive effects of learning computer programming. New Ideas in Psychology. Volume 2 pp 137-168.

PEERY, J. Craig and PEERY, I. W. (1987) The Role of Music in Child Development. In Peery, J. Craig, Peery, Irene Weiss, and Draper, T. (Eds). Music and Child Development. Springer-Verlag. p 15.

PIRTLE, M. and SEATON, K. P. (1975) Use of music training to actuate conceptual growth in neurologically handicapped children. Journal of Research in Music Education. Volume 24 (4) pp 292-301.

POLLARD-GOTT, L. (1983) Emergence of thematic concepts in repeated listening to music. Cognitive Psychology. Volume 15 (1) pp 66-94.

POSNER, M. I. (1973) Cognition: an introduction. Glenview. Scott, Forresman.

POSTMAN, L. (1971) Transfer, Interference and Forgetting. In Kling, L. W. and Riggs, L. A. (Eds) Experimental Psychology. Holt Rinehart and Winston pp 1019-1032.

PRESSLEY, M., SYNDER, B. L. and CARIGLIA-BULL, T. (1987) How can good strategy use be taught to children? In Cormier, S. M. and Hagman, J. D. (Eds). Transfer of Learning: contemporary research and application. Academic, Harcourt Brace Jovanovich. pp 81-120.

RAUSCHER, F. H., SHAW, G. L., KY, K. (1995) Listening to Mozart enhances spatial-temporal reasoning: towards a neuropsychological basis. Neuroscience Letters. Volume 185 pp 44-47. (also in Scientific Correspondence, Nature. 14 October 1993, p 611)

ROYER, J. M. (1979) Theories of the Transfer of Learning. Educational Psychologist. Volume 14 pp 53-69.

SCHENKER, H. (1906) Harmony. (Ed O. Jones and translated by E. M. Borgese) Cambridge: MIT Press, 1954.

SCHMIDT, R. A. (1975) A Schema theory of discrete motor skill learning. Psychological Review. Volume 82 pp 225-259.

SCHOENBERG, A. (1954) Structural Functions of Harmony. (Ed L. Stein) NY: Norton, 1969.

SHARMAN, E. (1981) The impact of music on the learning of young children. Bulletin of the Council for Research in Music Education. Volume 66 pp 80-85.

SHUELL, T. J. (1986) Cognitive conceptions of learning. Review of Educational Research. Volume 56 (4) pp 411-436.

SINGLEY, M. K. and ANDERSON, J. P. (1989) The Transfer of Cognitive Skills. Harvard University Press.

SLOBODA, J. (1985) The Musical Mind: the cognitive psychology of music. Oxford University Press.

SMYTH, M. M., MORRIS, P. E., LEVY, P. and ELLIS, A. W. (1987) Cognition in Action. Hillsdale NJ Lawrence Relbaum Association Limited.

SPYCHIGER, M., PATRY, J., LAUPER, G., ZIMMERMANN, E. and WEBER, E. (1995) Does more music teaching lead to a better social climate? In Olechowski, R and Svik, G. (Eds). Experimental Research in Teaching and Learning. Bern: Peter Lang.

THORNDIKE, E. L. and WOODWORTH, R. S. (1901) The influence or improvement in one mental function upon the efficiency of other functions. Psychological Review. Volume 8 pp 247-261.

THORNDIKE, E. L. (1913) Educational Psychology. NY Teachers College, Columbia University.

THORNDIKE, E. L. (1923) The effect of changed data upon reasoning. Journal of Experimental Psychology. Volume 5 pp 33-38.

THORNDYKE, P. and HAYES-ROTH, B. (1979) The use of schemata in the acquisition and transfer of knowledge. Cognitive Psychology. Volume 11 pp 82-106.

THUNE, L. E. (1950) The effect of different types of preliminary activities on subsequent learning of paired-associate material. Journal of Experimental Psychology. Volume 40 pp 423-438.

UNDERWOOD, G. and BRIGHT, J. E. H. (1996) Cognition with and without awareness. In Underwood, G. (Ed). Implicit Cognition. Oxford Science Publications.

WEBER, M. (1921) Die Rationalen und Sozialogischen Grundlagen der Musik. Drei Masken Verlag. Munich.

WEINSTEIN, C. E. and MAYER, R. E. (1985) The Teaching of Learning Strategies. In Wittrock, M. C. (Ed). Handbook of research on Teaching. (3rd Ed pp 315-327) MacMillan NY.

WITKIN, R. (1974) The Intelligence of Feeling. Heinmann.

WOLFF, K. L. (1978) The non-musical outcomes of music education: a review of the literature. Bulletin of the Council for Research in Music Education. Volume 55 pp 1-27.

ZULAUF, M. (1993) Three year experiment in extended music teaching in Switzerland: the different effects observed in a group of French speaking pupils. Quarterly Journal of Music Teaching and Learning. Volume iv (2).