Jump to ContentJump to Main Navigation
Musical ExcellenceStrategies and Techniques to Enhance Performance$

Aaron Williamon

Print publication date: 2004

Print ISBN-13: 9780198525356

Published to Oxford Scholarship Online: March 2012

DOI: 10.1093/acprof:oso/9780198525356.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 26 February 2017

Strategies for Memorizing Music

Strategies for Memorizing Music

(p.123) Chapter 7 Strategies for Memorizing Music
Musical Excellence


Oxford University Press

Abstract and Keywords

This chapter reviews a range of memorization strategies that are available to musicians. The chapter is in three main sections. It begins by setting out basic principles of human memory, including a discussion of the psychological structures thought to be involved in memorization and some recommendations for improving memory generally. The second section explores musical memory specifically and offers a selection of strategies commonly used by musicians and tested through systematic research. These include rote memorization, methods for memorizing visual and auditory information, methods for developing and exploiting ‘conceptual’ memory, and memorization for words and music. The chapter closes with a discussion of how these strategies can be used in conjunction with one another and how individual differences in learning styles may impact on one's capacity to memorize.

Keywords:   memorization techniques, musicians, human memory, musical memory, learning style

It is common in many kinds of music for artists to perform from memory. Within certain musical genres, in fact, social convention and audience expectation are such that performing with a notated score would be inconceivable. Rarely does one see a folk, rock, or pop musician reading from the printed page during performance; within the Western art music tradition (where audiences have come to expect almost note-perfect renditions of well-known pieces), memorized performances are customary for concert soloists. Why is this the case? Clearly, there are a number of practical advantages to playing or singing without a score. These include not having to turn pages, being able to monitor the physical aspects of a performance (e.g. looking at the hands or at other ensemble members), and in the case of concert pianists, avoiding the need to share the platform with a page-turner.

More importantly, perhaps, performing from memory is often seen to have the effect of enhancing musicality and musical communication. It is commonly argued that the very act of memorizing can guarantee a more thorough knowledge of and intimate connection with the music (see Plunkett Greene, 1912; Hughes, 1915; Matthay, 1926). In addition, memorization can enable the use of direct eye contact with an audience (and gaze, in other ways, for expressive purposes) that is more convincing than when referring to the score. Those who “possess” the music in this way often convey the impression that they are spontaneously and sincerely communicating from the heart, and indeed, contemporary evidence suggests that musicians who achieve this are likely to find their audiences more responsive (see Davidson, 1993, 1994; Williamon, 1999). Moreover, when performers receive and react to visual feedback from the audience, a performance can become truly interactive, involving genuine communication between all involved.

While performing from memory may have benefits for the performer and audience alike, the task of memorizing music and giving a memorized performance can be arduous. Memory failure—however momentary—can have catastrophic consequences for a performer's self-confidence. The musician does not, in fact, need to have experienced such failure to be frightened of forgetting, and this may well contribute to a more generalized performance anxiety. Furthermore, the fear of memory lapses is as common among experienced professional musicians as with novices—perhaps more so, as they feel they have more to lose. Artur Rubinstein, for example, “abandoned the concer t stage mainly on account of the fact that he could no longer depend on his memory in public performance” (Hughes, 1915, p. 603).

(p.124) This chapter reviews a range of memorization strategies that are available to musicians. An analysis of their efficacy is based on a critique of existing observational and empirical research, most of which stems from studies carried out within the context of Western classical music. Nevertheless, the benefits of each are presented so that they may be applied equally well to performance in other musical styles. The chapter is in three main sections. It begins by setting out basic principles of human memory, including a discussion of the psychological structures thought to be involved in memorization and some recommendations for improving memory generally. The second section explores musical memory specifically and offers a selection of strategies commonly used by musicians and tested through systematic research. These include rote memorization, methods for memorizing visual and auditory information, methods for developing and exploiting “conceptual” memory, and memorization for words and music. Finally, the chapter closes with a discussion of how these strategies can be used in conjunction with one another and how individual differences in learning styles may impact on one's capacity to memorize.

7.1 Memory: Structure and general recommendations

7.1.1 Structure

Before considering strategies that musicians can use to exploit their memory effectively and efficiently, it is important to place musical memory into a broader theoretical context. Memory can be conceptualized in a variety of ways. In Atkinson and Shiffrin's (1968) modal model, for example, memory consists of three stores, as depicted in Figure 7.1. The first is the sensory store, which has sub-stores for the different kinds of environmental information we receive via our senses: visual, auditory, tactile, and so on. Much of this incoming information is not attended to and, consequently, remains only briefly in memory. Information that does capture attention, however, is transferred into short-term memory (subsequent researchers have proposed that short-term memory hasa working memory component that is used to manipulate different kinds of information in conscious awareness; see Baddeley & Hitch, 1974). Memory storage at this level plays a vital role in basic, everyday functioning (e.g. remembering phone numbers and keeping track of an ongoing conversation). In terms of musical performance, this type of memory is integral to tasks such as sight-reading and improvisation, where new information must be retained and related to past, present, and predicted future events (see Chapter 8 for further details on sight-reading and improvisation, Chapter 11 for research on working memory enhancement, and Chapter 14 for sources of memory impairment).

Once in short-term memory, information can be rehearsed and elaborated in such a way that it is stored alongside or assimilated into existing knowledge. Unlike short-term memory, which has a capacity limited to around seven “chunks” of information (Miller, 1956), long-term memory seems to be of unlimited capacity and duration. It holds different kinds of information, including procedural knowledge (“knowing how” to do something), semantic knowledge (facts, i.e. “knowing that”), and episodic memories (details of events or episodes in one's life). These memories are organized (p.125)

                      Strategies for Memorizing Music

Figure 7.1 A graphical representation of the components of memory, consisting of (1) the sensory store, which temporarily holds environmental information received via the senses; (2) short-term memory, which contains information that captures attention and enables the manipulation of that information through working memory; and (3) long-term memory, which stores information that has been rehearsed and elaborated in such a way that it becomes integrated into one's knowledge. The last of these is particularly vital to the memorization of music and houses procedural and semantic knowledge (“knowing how” and “knowing that,” respectively), as well as episodic memories about events in one's life (diagram adapted from Baddeley, 1990; Eysenck & Keane, 2000).

hierarchically, enabling us to make associations both within and between different categories of information. Procedural knowledge underlies the performer's ability to make music, in terms of coordinating his or her movements to produce sequences of notes, such as scales and arpeggios, on an instrument. However, the knowledge that a particular sequence of notes represents certain patterns (a scale of D major, say, or a favorite melody) depends on semantic memory. Meanwhile, the specific associations (p.126) the melody has for the individual (e.g. where it was first heard, events going on at the same time as first hearing, others who were present) are held in episodic memory.

Of course, one of the primary objectives of storing information in long-term memory is to be able to retrieve that information accurately and on demand. Explicit retrieval of stored information takes place through the processes of recognition and recall. Recognition involves matching new, incoming information with information already stored in long-term memory. Recall involves actively (re-) producing that information. Both processes can occur spontaneously (the same cue can trigger recognition or recall), but recall makes greater demands on our ability to remember. The focus of this chapter is on the rehearsal and elaboration strategies that contribute to deliberate memorization in music: recall from long-term memory.

7.1.2 General recommendations

While all people possess the same memory structures and use the same basic cognitive mechanisms, some people seem to have “better” memories than others. To a limited extent, this may be due to differences in “natural” general memory ability. However, Wilding and Valentine 1997 conclude from their research with exceptional memorizers that most superior memory performance can be attributed to the use of highly effective strategies for learning (or “encoding”) and, independently, for retaining information in memory.

There are many accounts of professional musicians exhibiting extraordinary feats of memory (Revesz, 1925/1999; Marcus, 1979; Marek, 1982; Charness et al., 1988). The notion of individual differences in the ability to memorize music, even among those who achieve such feats, is supported by research with musicians memorizing music in the laboratory (Lehmann & Ericsson, 1995; Ginsborg, 2000) and under more naturalistic circumstances in the studio (Ginsborg, 2002). Meanwhile, everyone who plays or sings, no matter what their level of natural ability, can benefit from learning and practicing effective strategies for memorizing music.

Before considering the strategies that have been found to be particularly useful for memorizing music, there are general recommendations for encoding information and storing it in long-term memory for easy retrieval that can aid the musician. These can be categorized in four ways: (1) improving memory in its most general sense; (2) staving off or remediating age-related memory deterioration; (3) enhancing study skills; and (4) using mnemonics by associating meaningless information with the material to be remembered. Firstly, recommendations for improving memory in its most general sense include understanding how memory works and having the motivation to try to remember (Arden, 2002). Secondly, those who fear age-related memory deterioration are advised to reduce stress, eat a balanced diet, take vitamin supplements, avoid alcohol and drugs, keep an active mind, and do “brain exercises” (Small, 2003; see also Chapters 3, 9, 10, 12, and 14 of this volume). Thirdly, students can improve their learning, and specifically their performance in examinations, by using memory strategies such as attending to the material to be remembered (e.g. in lectures) before organizing, interpreting, and understanding it, and also by visualizing, reciting, and reviewing it (Pauk, 1994). Different techniques focus on enhancing different aspects of memory. For example, “mind-mapping” (Buzan, 1989) involves organizing (p.127) and making appropriate links between chunks of information. The end-product (the “mind-map”) provides an elaborated set of notes; more importantly, the process of devising an idiosyncratic visual representation serves to reinforce memory. Finally, mnemonics include “peg-words,” such as rhyming words and numbers (one = bun, two = shoe); the method of loci, dating from classical times, whereby objects or concepts are associated with familiar locations such as landmarks on a regular journey, or different rooms in one's home; and linking unrelated items by weaving them into a story (Arden, 2002).

Many of these recommendations, such as those relating to health and well-being, are supported by continuing research into the functioning of the brain (Small, 2003). Understanding how memory works may seem an ambitious aspiration, yet it is well known that metacognition, or “knowing about knowing,” is essential to learning (see Chapter 5 for specific recommendations for individual practice). When students are advised to attend to and later review what has been said in lectures, this is based on the notion of transferring information from the sensory store into short-term memory and thence into long-term memory. Interpreting new information in order to understand it, and linking it with information already stored in long-term memory is a highly effective way of extending the storage capacity of short-term memory by increasing the size of information chunks. This was shown dramatically in a study carried out by Ericsson et al. (1980), in which one of the authors, a keen amateur runner, learned to encode up to 82 digits by thinking of them as combinations of running times (e.g. 3594 was stored as 3 minutes 59.4 seconds for a mile run, which was meaningfully classified as “fast”). Visualization and recitation are forms of elaboration and rehearsal, drawing on visual, kinesthetic, and auditory processes; reviewing involves reorganization of the material to be remembered in long-term memory. Similarly, mind-mapping combines the creative use of visual and associative memory. As for the mnemonics outlined above, they all rely on forming associations between items, using visual and/or auditory memory, in order to create new meanings and links between the new information to be remembered and existing information stored in long-term memory. Still, how do these recommendations relate to memorizing music?

  • You are well advised to take care of your general health for sharp and long-lasting memory performance.

  • One of the main aims of this chapter is to help you find out what is currently known about how memory works and how musicians go about memorizing music so that you can learn to use effective memorizing strategies yourself. These are based on similar principles to those that underpin most study skills.

  • However, while study skills usually relate to the learning and memorization of semantic knowledge, you need procedural knowledge as well to sing or play. This is gained through practice involving some inevitable repetition, in order to “automatize” performance.

  • As for mnemonics, they may not at first glance seem relevant to music memorization. Yet, they can be useful for encoding sequences of musical events within a piece, or the words of songs that do not seem to make sense otherwise.

(p.128) 7.2 Musical Memory

One way of describing a memory for a specific piece of music—whether acquired through repeated hearings or deliberate memorization—is a “mental representation,” stored in long-term memory, on which the musician can draw when performing. Lehmann (1997) suggests a structure for expert musicians' mental representations with three facets: (1) the desired performance goal, (2) the knowledge of how to produce that goal, and (3) the actual performance. The desired performance goal represents an ideal performance: a rehearsed, highly detailed interpretation of the music. This interpretation can be conceptualized at different levels, forming a hierarchy (Williamon et al., 2002). At the most global level, the music is represented mentally as a whole piece, from beginning to end; at the most local level, it is represented by the notes forming each musical sequence. Of course, there are numerous possible hierarchical levels in between. In a classical sonata movement, these might be the exposition, development, and recapitulation; in a song they might be introduction, verse, chorus, and so on. The desired performance goal is achieved (to a greater or lesser extent) via the second and third facets of the mental representation suggested by Lehmann—that is, the production aspects and the actual performance. The representation of production aspects is constructed from repeated rehearsals of musical sequences that become fully automated but which must continually be monitored. The representation of the actual performance—the interpretation of the music—also involves monitoring in the context of the performance situation and feedback from the audience. In terms of the different kinds of long-term memory outlined in Section 7.1.1, musicians' semantic and episodic memories for the music they perform inform the desired performance goal. Meanwhile, actually playing, singing, or conducting music requires procedural memory.

Different kinds of memorizing strategies can contribute to the formation of mental representations at their different levels and enable attention to be shifted, during practice, from one level to another (Williamon et al., 2002). Whether the information coming into the sensory store is tactile, auditory, and/or visual (e.g. the feel of the fingers on the keys of the instrument, the sound of the music, and the look of the musical score, respectively), sensory information is essential for mental representations of music at the local level. It is also essential for producing and monitoring the desired performance goal. Sequences of musical events are usually memorized—and automatized—via rote repetition using “kinesthetic” memory. However, they could not be learned in the first place without the use of “visual” memory of the notated music and of the way the music is created, and “aural” memory of sounds that are imagined, heard, or produced by the performer for the first time. Depending on the learning preferences of the musicians, music can thus be rehearsed and elaborated kinesthetically, visually, and aurally for representation in long-term memory.

However, in order to represent music mentally at global and local levels and to move between the different levels of representation during practice, analytic or conceptual strategies are required. These strategies have already been introduced in relation to study skills; they enable us to understand and organize the material (in this case, the music) to be remembered. Unlike kinesthetic, visual, and aural representations, (p.129) conceptual memory for music can be thought of as a form of semantic knowledge (as well as procedural), which facilitates performance.

So how do musicians go about memorizing using these different strategies? Some memorize, apparently, almost without conscious awareness. When asked how they memorize they cannot—or will not—say. Many musicians, on the other hand, have shared their experiences of memorizing, in published books and interviews, passing on the techniques they have found useful to their students. The list that follows is based on these, as well as on a growing body of research in psychology of music. Of course, no single strategy is guaranteed for all musicians and individual differences in learning styles are discussed in Section 7.3.

7.2.1 Memorizing by rote

Rote memorization is perhaps the most common form of rehearsal, particularly for novice musicians attempting to commit a piece of music to memory. Essentially, it uses kinesthetic memory (otherwise known as tactile, motor, finger or muscular memory) and consists of repetition of a bar, phrase, or page until it can be played, automatically, by “feel.” For example, many pianists report memorizing with separate hands and consistently using the same fingerings so as to develop their kinesthetic memory for a piece of music (Chaffin et al., 2002). In a study of singers memorizing words and music separately and together, however, the latter performed with greater accuracy and confidence, since the words and music acted as cues for each other (Ginsborg, 2000).

◆ If you are a keyboard player, you could memorize by rote with your hands separately as well as together; for singers, try to use it to link the words and music together.

Expert musicians use rote memorization to “overlearn” their performances, to ensure that they will be as secure as possible, but they almost always use other memorizing strategies in addition to rote memorization. This is because kinesthetic memory is vulnerable to interference. A stray thought, a cough from the audience, hearing something in the music, as it were, for the first time can “throw” the performer.

◆You should use rote memorization in order to free your mind to monitor your performance, to focus on the meaning you wish to convey, and to communicate with your fellow musicians and/or your audience.

In an interview study carried out by Hallam (1997) with novice musicians and professional orchestral musicians, the majority reported memorizing—initially—without conscious awareness, as they learned their parts. When they identified passages they could not play accurately from memory, they chunked them into smaller sections and repeated each section until it was secure. Then, they grouped the small sections into longer sections, and repeated these longer sections until they too were secure. In this way, the whole piece would eventually be successfully memorized. This technique was useful for short pieces; however, for longer, more complex works, half of the professional musicians in Hallam's sample reported using what she refers to as “cognitive analysis” (see Section 7.2.5) to construct a framework into which the memorized sections could be placed.

(p.130) 7.2.2 Memorizing visual information

Most classical musicians begin learning and practicing new pieces from a notated source, and this can form an important foundation for memory. Indeed, many musicians report “knowing where they are on the page” when they play or sing notated music from memory—a very useful tool for preventing or recovering from memory lapses. As early as 1800, long before it was customary for classical musicians to perform from memory, Anfossi (cited in Monahan, 1978) recommended that singers should memorize at least the turn of the page.

◆ It is not a good idea to rely simply on the visual memory you acquire in the course of learning the music. This is similar to memorizing without conscious awareness; like rote memorization, it can be an unreliable technique on its own. Rather, use memory for visual information deliberately; you could mark the score with a pencil or colored pen.

Although there is anecdotal evidence of musicians' (particularly conductors') memory for musical scores (see Marcus, 1979), empirical research on memory purely for visual information is scant. Nuki 1984 asked piano and composition students first to sight-read and then to memorize a novel piece of music. When they had memorized it, they performed it from memory. They were then asked which of four strategies they had used to memorize: kinesthetic, visual, aural, or a mixture of the three. The students who reported using a visual strategy were significantly quicker to memorize than those who used one of the other strategies or a mixture of the three. However, by “visual” strategy, Nuki infers the use of the “inner ear” (see Section 7.2.3), and indeed, the quickest memorizers in her sample were the students of piano and composition who were also good sight-readers and had gained high grades in solfège at their universities.

◆ Deliberate memorization of visual information, whether signals from another musician or the score itself, is linked to conceptual memory see Section 7.2.5. You should carry out some form of analysis of the music in order to decide which information will serve as the most reliable cues and then mark those cues in the score. Each time you rehearse the piece thereafter, you are reinforcing (sometimes unknowingly) your memory of those cues.

Visual memories can also consist of visions of the position of the instrument and of the body. Where musicians play in groups they often rely on each other's visual, as well as aural, cues.

◆ If you are playing or singing with an ensemble, visual memory may be triggered by something as basic as a conductor's down-beat or a particular movement by a section leader. These are not just trivial remembrances, but rather they can play a major role in the unfolding of a performance. For maximal efficacy in practice, particularly useful cues as such should be explicitly acknowledged to ensure their consistent delivery.

7.2.3 Memorizing by ear

Many musicians—particularly musicians who are not part of the Western classical tradition—do not rely on musical notation to learn or memorize, but rather memorize (p.131) through listening and imitating what they hear. For some, this is laborious and time-consuming; for others, comparatively easy. Macek 1987 reports the case of a jazz musician who developed what he called a “photographic” ear (analogous to eidetic memory for visually-presented stimuli; Haber & Haber, 1988), which enabled him to recall and reproduce complex music in detail. Aural memory is as hard as visual memory to disentangle from the formation of mental representations and conceptual memory. Two studies in particular illustrate this. First, Sloboda et al. 1985 studied the exceptional verbatim memory for music of an autistic savant and showed that his memory was reliable only when he listened to music in a familiar genre. Second, Sloboda and Parker (1985) asked musicians and nonmusicians to listen to short folk melodies and then sing them as accurately as possible from memory. On the whole, the versions produced were inaccurate and/or incomplete, but they preserved the metrical and—particularly in the case of the musicians—harmonic structure of the originals. This suggests that we construct memory for specific pieces of music using what we already know about and have experienced with music; every new piece of music that we hear contributes to this knowledge.

  • While some musicians rely on the aural memories they acquire in the course of learning music, this is rarely reliable without conceptual memory to serve as a framework for monitoring “where you are” in performance and “where you are going,” and to provide you with a repertory of alternative musical sequences with which to improvise, should you have a memory lapse.

  • Once you have a conceptual framework, using aural memory as a basis for rote repetition, either playing or singing aloud, or in your imagination, is an excellent strategy.

7.2.4 Translating visual information into imaginary sound

An important tool for musicians who do use musical notation is the ability to imagine the way a performance would sound, or how it would feel to play or sing, from the score. Gieseking and Leimer (1932/1972) call this “visualization” and recommend that pianists should memorize, referring to the printed score, using their “inner ear” before beginning to practice at the keyboard. Gordon (1993) refers to this as “audiation”: being able to hear and understand music that is no longer, or may indeed never have been, audible. Equally, it is possible for musicians who do not use musical notation to hear a piece of music and imagine the way they, or someone else, would perform it. The research by Nuki outlined in Section 7.2.2 highlights the role of the ability to read music, translating visual information into imaginary sound and thus creating a mental representation for music.

◆ The difference between reading music and playing or singing aloud, and reading music and conjuring up imaginary sound (and “feel”) is like the difference between reading a piece of text aloud, as beginners do, and reading silently. If you are a musician who must learn and memorize notated music, then learning to hear it in your mind as you read it—so that you can memorize away from your instrument—is one of the most valuable skills you can develop.

(p.132) 7.2.5 Developing and exploiting conceptual memory

The ability to form mental representations of music at the global and local levels, particularly for extended compositions, depends on “conceptual” memory: the musician's existing semantic knowledge—held in long-term memory—of the structures that underlie the music. These structures may vary according to musical genre, and they include chord progressions in jazz, rhythmic patterns, ragas in Indian music, and the conventions of tonality in Western classical music.

Lehmann and Ericsson 1995 tested the hypothesis that the ability to construct and draw on a mental representation facilitates performance from memory that is both stable (i.e. the musician can reliably “keep the music the same”) and flexible (i.e. variations can be made in performance, depending on the requirements of the situation). They asked advanced students of piano to memorize two short pieces of music, providing “think-aloud” verbal reports as they did so. The pianists memorized the pieces—accompaniments to an instrumental “pacing” track—in the laboratory by carrying out a number of pairs of trials, playing each piece once with the musical score and once without until they could perform it from memory without errors twice, consecutively: the fewer trials they needed, the faster they had succeeded in memorizing. Then they carried out a series of tasks designed to test their ability to “manipulate” the pieces from memory. The results of the study showed that the pianists who were quickest to memorize made the fewest errors when transposing from memory. Thus, rather than learning musical sequences by rote and reproducing them using kinesthetic memory, it is clear that efficient memorization must involve the ability to construct and draw on a detailed mental representation of the music, allowing the performer both to reproduce it reliably and to manipulate it in performance from memory.

Ginsborg (2000) carried out a similar study to investigate the nature of mental representations for the words and music of songs with 20 singers. Each singer memorized two songs, in two separate sessions, paced by a recorded accompaniment, using the Lehmann and Ericsson 1995 memorizing paradigm. Once the songs were memorized, the singers carried out a number of tasks. Transposition—the task the quickest memorizers in Lehmann and Ericsson's study performed most accurately—is relatively simple for most singers, so instead they were asked to retrieve the words and melody of the song independently, to “deconstruct” the words and melody, to show an awareness of the compositional structure of the song, to use visual memory to retrieve information about different aspects of the musical score, and to respond to cues taken from the words and music of the song.

There was no correlation between speed of memorization and performance on some of the more straightforward tasks, such as performing the song accurately with accompaniment and at the same tempo as it was originally memorized. This would suggest a role for the automatization of performance, through the rote repetition of text and melody together (see also Section 7.2.6). On the other hand, speed of memorization was significantly correlated with performance on the seven most difficult tasks: speaking the words of the whole song at speed; retrieving fragments of the text and fragments of the music, with the words and pitches in reverse order; responding to cues, both forward and reverse; singing the phrases of the song in reverse order; and singing the pitches of the melody of the song only, without rhythm, to the regular beat (p.133) of a metronome. Such tasks could not have been performed successfully in the absence of a well-formed, multicode mental representation of the song including conceptual, kinesthetic, visual, and aural information. As in Lehmann and Ericsson's study, the faster the singers acquired these mental representations, the better they performed the tasks. Mental representations of music must, therefore, underlie the ability to produce performances from memory that are both stable and flexible.

◆ Once you have memorized a piece for the first time, you should test your memory in order to simulate different possibilities for recovering from lapses. You could try starting mid-section instead of at a structural boundary, for example, or by practicing jumping from one section to a previous section and to later sections. When you have done this, you will know you have created a mental representation on which you can draw to produce a stable, yet flexible, performance.

Research has focused on two main ways of developing and exploiting mental representations using conceptual memory: analysis and the use of structural boundaries to organize practice. These are discussed and recommendations are offered in turn.


It is often suggested that musicians should analyze the music they are to memorize (e.g. Hughes, 1915; Matthay, 1926), partly so as to be able to “chunk” or to divide the music into shorter sections; once each one is memorized they can be “grouped” together. Many researchers have investigated the efficacy of analysis as a memorization strategy. For example, half the professional musicians—though none of the novice musicians—who took part in Hallam's (1997) study reported using an analytic approach to develop a framework into which they could fit the material they had memorized by rote. According to Hallam, this included noting where the music changed from one key to another, particular harmonic progressions, the length of rests, and difficult exit points.

Rubin-Rabson 1937, on the other hand, compared four groups of piano students' memorization of short pieces. One group memorized at the keyboard. Two groups analyzed the pieces before memorizing them, one independently, the other with Rubin-Rabson's guidance. The fourth group listened to a recording of the pieces and then memorized in one of the three ways already detailed. Three weeks later, the students who had either analyzed the pieces themselves or with guidance were the quickest to relearn the pieces and to perform them accurately from memory.

Ross 1964 replicated this study with the participation of competent wind players randomly divided into experimental and control groups. Following a pre-test to establish baseline memorizing ability, the experimental group undertook a weekly hourlong session with a teacher, for 6 weeks, in which they analyzed and memorized a series of training pieces, in preparation for memorizing a test piece at the end of the 6-week period. Again, the experimental group accurately memorized the test piece significantly faster than the control group.

The most detailed account of the use of analysis to develop conceptual memory for music is provided by Chaffin and Imreh who have been collaborating for almost a decade on a project exploring the cognitive processes involved in learning and memorizing (p.134) music (see Chaffin et al., 2002). The purpose of their first study was to produce a systematic record of the activities of an experienced professional concert pianist (i.e. Imreh) while memorizing the third movement of Bach's Italian Concerto (Chaffin & Imreh, 1994). Imreh undertook 57 practice sessions over the course of 10 months, 34 of which were recorded on videotape and 11 on audiotape. She talked aloud about her decision-making as she worked; she was interviewed and, once the piece was successfully memorized, she commented on her videotaped practice and annotated the score to show how she organized the music conceptually. One of the findings of this study was that she identified the compositional structure of the piece and used it not only to guide her practice, through chunking and grouping, but also to ensure accuracy when she played from memory. She practiced and memorized acknowledged sections in the music separately and together. When she came to play the whole piece from memory, she would deliberately recall the particular features of each section in order to remind herself which section she was actually playing and which came next (Chaffin & Imreh, 1997).

  • The use of conceptual memory is the crucial overarching strategy that no musician can do without.

  • Whatever kind of music you play or sing, you need to identify how it fits together (i.e. its structure), whether this is at a highly detailed level such as that described above or at a much simpler level—remembering, for example, that you must sing two verses interspersed with three choruses, the last time in a higher key.

  • How you go about identifying structure and using it to help you memorize will depend on the nature of your musical training as well as the nature of the music. Some musicians rely on an understanding of underlying harmony; others rely on memory for melodic sequences; others rely on identifying and being able to reproduce rhythmic patterns. In Western classical music, all of these are useful ways of conceptualizing music, separately and together.

Structural boundaries

Williamon and Valentine 2002 explored the development of the ability to identify structure, to use it as a tool for organizing practice, and to employ it successfully for retrieval in performance. They asked 22 pianists to learn and memorize a piece by J. S. Bach appropriate to their level of expertise. All of the pianists' practice sessions were recorded on audiotape. The final performances that they gave from memory, as part of a recital, were recorded on videotape and assessed by experienced adjudicators. The pianists made verbal and written comments on their practice, annotated their musical scores and were interviewed following their final performances about the practice and memorizing process. The results of the study showed that pianists with higher levels of skill were more likely than those at lower levels to practice segments of the piece beginning and ending on bars at the structural boundaries (these were defined according to the structure of the piece as understood and described by the individual pianist; only three pianists' descriptions conformed to the “textbook” formal structure of the piece). The likelihood that a pianist would use structural boundaries to guide his or her choice of the part of the piece to be practiced increased over the course of practice sessions. (p.135) The earlier the pianist began to do so, however, the better his or her final performance was likely to be judged.

Research has shown that singers, too, use structural boundaries to guide their practice during the course of memorization. Ginsborg 2002 carried out an observational study in which 13 singers learned and memorized the same song over the course of up to six 15-minute practice sessions. The sessions, which included concurrent verbal commentaries made by the singers, were taped and transcribed. Finally, an assessment of the singers' speed and accuracy of memorization was made in order to determine which of the various memorizing strategies used by the singers were more and less effective. Fast, accurate memorizers were no more likely than slower, less accurate memorizers to use structural boundaries to guide their practice. Furthermore, the use of structural boundaries did not change significantly between the beginning and end of the practice sessions. However, it was used widely: more than four fifths of “attempts”, the segments of the song that were practiced, began on one of its eight phrases. Nearly half of the attempts began at the start of a phrase and finished at the end of the same or a subsequent phrase. The majority of attempts were on the first phrase, followed closely by the third phrase (i.e. the beginning and middle of the first verse); a much smaller proportion of attempts was made on the fifth and seventh phrases (the beginning and middle of the second verse). An unexpectedly high proportion of attempts on the final phrase was attributable to the fact that participants found it “technically” difficult to sing (it had two large-interval “leaps”) and also to memorize.

  • You should try to divide the music into sections in a way that makes sense to you, so as to memorize the sections one at a time before re-combining them into longer sections and fitting them into your conceptual framework for the whole piece.

  • The more you organize your practice around sections identified in the music (especially in the early stages of learning a piece), the more likely you are to remember this basic architecture in performance.

7.2.6 Memorizing strategies for singers: combining words and music

Like instrumental musicians, singers use kinesthetic, visual, aural, and conceptual memory. Unlike instrumental musicians, however, they must also memorize words. It should be remembered that music and words are not necessarily equal partners: sometimes the voice is used instrumentally (e.g. scat singing in jazz or choral music by Bach) and sometimes the words are more important than the melody (e.g. operatic recitative, Sprechstimme). The relative importance of words and music in a particular musical genre will determine the memorizing strategies used, and whether the words and music are memorized separately or together. In an observational study, Ginsborg 2002 identified a variety of “modes” of attempt used in singers' practice sessions. These were: focusing on the words alone (speaking the words, while reading from the score and from memory); focusing on the music alone (playing the melody, playing the accompaniment, vocalizing the melody, and counting—the last of these while reading from the score and from memory); and singing the words and music together, while reading from the score and from memory. Fast, accurate memorizers used more modes of attempt than the slower, less accurate memorizers. They were also more likely to (p.136) count aloud, but most importantly, they began memorizing earlier and sang more of the song from memory during the course of their practice sessions.

Next, the higher-level strategies of the “best” memorizer (the first of the 13 participants in the sample to sing the whole song entirely accurately from memory) and the “worst” memorizer (the one who took longest to memorize and made the most errors when she sang from memory) were compared, with reference to their verbal commentaries as well as their practice data. The “best” memorizer sang the words and music of the song together rather than separately. She started memorizing early and tested her memory throughout the practice sessions. She worked on a variety of lengths of practice unit. She made plans and implemented them, monitored and corrected her errors, and explicitly evaluated her practice. Her strategies were varied and complex; furthermore, her verbal commentary was detailed and self-guiding. In contrast, the “worst” memorizer implemented plans, monitored errors, and evaluated her practice to a much lesser extent. She preferred to sing the music only, started to memorize comparatively late and consistently went back to the beginning of the song and repeated it in its entirety, rather than dividing it into smaller sections.

Ginsborg (2000) also investigated which of two strategies is more effective for memorizing a song: memorizing the words and melody separately or memorizing them together. Sixty singers took part in this study. Each singer was categorized in two ways: by their experience of memorizing and by their level of musical expertise. Half of them were solo singers with considerable experience of memorizing and performing from memory, while the other half were choral singers with little or no experience of memorizing songs. Half of them had undertaken musical training to advanced levels, while the other half had lower levels of musical training.

They were randomly divided into three groups and asked to memorize a short, unaccompanied song in a single 20-minute practice session. The first group memorized the words first, then the melody, and finally both together; the second group memorized the melody first, then the words, and finally both together; the third group memorized the words and melody simultaneously throughout the course of the practice session. When the singers had memorized the song, they were asked to sing it from memory twice, their two performances separated by a 10-minute interview. Their second performances from memory were transcribed, coded, and analyzed. Errors and hesitations were noted and counted. Generally, the singers who memorized the words and melody of the song together gave more accurate and more confident performances from memory. They made fewer word-only errors, and fewer simultaneous word-and-music errors. They also made fewer hesitations. So, as shown in the comparison of the strategies used by the “best” and “worst” memorizer in the observational study, memorizing words and melody together appears to be a more effective strategy than memorizing them separately.

A comparison of the performances of singers with high and low levels of memorizing experience, and with high and low levels of musical expertise, was also instructive. Those singers with considerable experience of memorizing songs were no more accurate or confident in their performances from memory than those with no experience of memorizing songs. In other words, experience is no guarantee of success, at least in (p.137) terms of memorizing. On the other hand, there were observable differences between those singers who had advanced levels of musical training and those who did not. The singers with high levels of musical expertise made fewer music-only errors, and fewer simultaneous word-and-music errors, than the nonexpert singers. Furthermore, while—as one might predict—they were generally more accurate in their performances from memory, this was only the case when they had memorized the words and melody together.

  • Almost every memorizing strategy recommended in this chapter is as applicable to singers as it is to instrumental musicians. The only difference between them is that singers must memorize words as well as music. Think of the words as an element of the music rather than the other way round.

  • You might think of the words in terms of the way they look on the page (visual memory), the way they sound (aural memory), or the way it feels to produce them (kinesthetic memory). If you are learning a melody using visualization/audiation, mouth the words to yourself. Once they are securely memorized, it is helpful, of course, to think about what they mean.

7.3 Conclusions

The aim of this chapter has been to report on memorizing strategies that have been shown to be effective for all musicians, both instrumentalists and singers. To reiterate: music is stored using multiple representations in long-term memory. Effective memorization strategies capitalize on this.

  • Analyze the music you are to memorize in order to understand it and organize it, using conceptual memory as an overarching framework.

  • Chunk the music and practice it in small sections, from one structural boundary to the next, increasing the size of chunk as you become more familiar with the music.

  • Use rote, kinesthetic memory in conjunction with either visual or aural memory, or both.

For the most part, discussion has focused on memorization of a specific piece of music; however, some final, more general suggestions can be made:

  • First, it has been shown in a number of domains that “little and often” is a better strategy for memorizing than trying to memorize in just one or two extended sessions. Sometimes you find that your memory for a piece of music has improved between practice sessions without conscious effort; alternatively, you can use the time between practice sessions for recalling the music you have memorized and identifying sections that may need more work.

  • Second, there are strategies that can be used in the final stages of preparing to give a performance from memory. One is state-dependence: learning to control your frame of mind so that you can conjure up the same sense of exhilaration, for instance, when you are practicing as when you are actually performing. The other is contextdependent learning: practicing in the room or on the stage where the performance is (p.138) to take place. These strategies, like so many others, can be useful whatever kind of music you are playing or singing, and whether you are a soloist or in an ensemble.

In addition, we have to remind ourselves that different tasks may require different strategies. Furthermore, we also need to remember that different individuals have different capacities for memorizing easily and successfully. While most research investigating memorization strategies has focused on the average numbers of trials required to memorize and relearn pieces of music, these mask a wide variety of memorization speeds. For example, Lehmann and Ericsson 1995 found that their slowest memorizer needed 6–7 times as many trials to memorize as their fastest memorizer; using the same paradigm, the singers in Ginsborg's (2000) study required from 2 to 17 trials to memorize a song with a word-text, and from 3 to as many as 49 trials to memorize the song with a digit-text. Similarly, in Ginsborg 2002, in which 13 singers learned and memorized under more (although not completely) naturalistic circumstances, the fastest memorizer produced 3232 beats before her first accurate performance from memory (equivalent to just under 20 renditions of the song) while the slowest produced 5287 beats (equivalent to 32 renditions).

These findings may well reflect the extent to which the musicians who took part in these studies found that the strategies they were encouraged to use resembled or interfered with their usual, or preferred, strategies. From interview and experimental data, it is clear that musicians not only have different memorizing abilities, but also a wide variety of views on memory and preferences for using different approaches to memorization (e.g. kinesthetic, visual, and aural). These preferences have been characterized by some educationists as learning modalities or learning styles, which can be identified through the use of checklists such as the Swassing-Barbe Modality Index (SBMI; Barbe & Swassing, 1979) and the Learning Styles Index (LSI; Price & Dunn, 1997; Dunn et al., 2003). Although learning style theories have been criticized on the grounds that they lack a psychological foundation (Sternberg & Grigorenko, 2001; Renzulli & Dai, 2001), some music teachers advocate teaching their students according to their particular learning strengths (see Garcia, 2002).

While the impact of learning styles on expert musicians' memorizing abilities and strategies has not yet been explored, research has been carried out using the SBMI and LSI with children: (1) to identify “gifted” musicians and investigate the extent to which their learning styles differ from those of the general population (Kreitner, 1981); (2) to find out if memory for rhythm patterns can be enhanced by presenting them in children's preferred modalities (Persellin & Pierce, 1988); (3) to examine the relationships between the perceptual elements of learning style, musical aptitude, and attitudes to music learning (Faulkner, 1994); and (4) to examine the influence of children's preferred modalities on the way they listen to music. The findings of these studies indicate that children's preferred modalities may influence some aspects of basic music listening and learning. However, there is only anecdotal evidence, as yet, to suggest that these persist into adulthood. If individual musicians do possess particular learning strengths then they must also have complementary learning weaknesses. Future research might well seek to determine the nature of the training whereby such weaknesses in memorization can be systematically improved.

(p.139) Further information and reading

Bibliography references:

Chaffin, R., Imreh, G., & Crawford, M. (2002). Practicing Perfection: Memory and Piano Performance. Mahwah, NJ: Erlbaum.

Gieseking, W., & Leimer, K. (1932/1972). Piano Technique. New York: Dover. (Original work published in 1932.)

Lehmann, A. C. (1997). Acquired mental representations in music performance: Anecdotal and preliminary empirical evidence. In H. Jørgensen & A. C. Lehmann (Eds.), Does Practice Make Perfect? Current Theory and Research on Instrumental Music Practice (pp. 141–163). Oslo: Norwegian Academy of Music.

Wilding, J., & Valentine, E. (1997). Superior Memory. Hove, UK: Psychology Press.

Williamon, A. (2002). Memorising music. In J. Rink (Ed.), Musical Performance: A Guide to Understanding (pp. 113–126). Cambridge: Cambridge University Press.


Bibliography references:

Arden, J. B. (2002). Improving Your Memory for Dummies. New York: Wiley.

Atkinson, R. C., & Shiffrin, R. M. (1968). Human memory: A proposed system and its control processes. In K. W. Spence & J. T. Spence (Eds.), The Psychology of Learning and Motivation: Advances in Research and Theory (Vol. 2, pp. 89–195). London: Academic Press.

Baddeley, A. D. (1990). Human Memory: Theory and Practice. Boston: Allyn and Bacon.

Baddeley, A. D., & Hitch, G. J. (1974). Working memory. In G. H. Bower (Ed.), The Psychology of Learning and Motivation (Vol. 8, pp. 47–90). London: Academic Press.

Barbe, W. B., & Swassing, R. H. (1979). Teaching through Modality Strengths: Concepts and Practices. Columbus, OH: Zaner-Bloser.

Buzan, T. (1989). Use Your Head. London: BBC Consumer Publishing.

Chaffin, R., & Imreh, G. (1994). Memorizing for Performance: A Case Study of Expert Memory. Paper presented at the Third Practical Aspects of Memory Conference, University of Maryland.

Chaffin, R., & Imreh, G. (1997). “Pulling teeth and torture”: musical memory and problem solving. Thinking and Reasoning, 3, 315–336.

Chaffin, R., Imreh, G., & Crawford, M. (2002). Practicing Perfection: Memory and Piano Performance. Mahwah, NJ: Erlbaum.

Charness, N., Clifton, J., & MacDonald, L. (1988). Case study of a musical “mono-savant”: A cognitive-psychological focus. In L. K. Obler & D. Fein (Eds.), The Exceptional Brain: Neuropsychology of Talent and Special Abilities (pp. 277–293). London: Guilford Press.

Davidson, J. W. (1993). Visual perception of performance manner in the movements of solo musicians. Psychology of Music, 21, 103–113.

Davidson, J. W. (1994). Which areas of a pianist's body convey information about expressive intention to an audience? Journal of Human Movement Studies, 26, 279–301.

Dunn, R., Dunn, K., & Price, G. E. (2003). Learning Style Inventory (LSI) Grades 5–12. Lawrence, KS: Price Systems.

Ericsson, K. A., Chase, W. G., & Falloon, S. (1980). Acquisition of memory skill. Science, 208, 1181–1182.

(p.140) Eysenck, M. W., & Keane, M. T. (2000). Cognitive Psychology: A Student's Handbook (4th ed.). Hove, UK: Psychology Press.

Faulkner, D. L. (1994). An Investigation of Modality Preferences, Musical Aptitude, and Attitude Toward Music at the Third-Grade Level. Unpublished doctoral dissertation, University of Mississippi, Oxford, MS.

Garcia, S. (2002). Learning styles and piano teaching. Piano Pedagogy Forum, 5(1). Retrieved August 28, 2003, from www.music.sc.edu/ea/keyboard/PPF/5.1/5.1.PPFpp.html.

Gieseking, W., & Leimer, K. (1932/1972). Piano Technique. New York: Dover. (Original work published in 1932.)

Ginsborg, J. (2000). Off by heart: Expert singers' memorisation strategies and recall for the words and music of songs. In C. Woods, G. Luck, R. Brochard, F. Seddon, & J. A. Sloboda (Eds.), Proceedings of the Sixth International Conference on Music Perception and Cognition. Keele, UK: Keele University.

Ginsborg, J. (2002). Classical singers learning and memorising a new song: An observational study. Psychology of Music, 30, 58–101.

Gordon, E. E. (1993). Learning Sequences in Music: Skills, Contents, and Patterns: A Music Learning Theory. Chicago: GIA Publications.

Haber, R. N., & Haber, L. R. (1988). The chacteristics of eidetic imagery. In D. Fein & L. K. Obler (Eds.), The Exceptional Brain: Neuropsychology of Talent and Special Abilities (pp. 218–241). London: Guilford Press.

Hallam, S. (1997). The development of memorisation strategies in musicians. British Journal of Music Education, 14, 87–97.

Hughes, E. (1915). Musical memory in piano playing and piano study. Musical Quarterly, 1, 592–603.

Kreitner, K. R. (1981). Modality Strengths and Learning Styles of Musically Talented High School Students. Unpublished masters dissertation, Ohio State University, Columbus, OH.

Lehmann, A. C. (1997). Acquired mental representations in music performance: Anecdotal and preliminary empirical evidence. In H. Jørgensen & A. C. Lehmann (Eds.), Does Practice Make Perfect? Current Theory and Research on Instrumental Music Practice (pp. 141–163). Oslo: Norwegian Academy of Music.

Lehmann, A. C., & Ericsson, K. A. (1995). Expert pianists' mental representation of memorized music. Poster presented at the 36th Annual Meeting of the Psychonomic Society, Los Angeles.

Macek, K. (1987). The photographic ear. Piano Quarterly, 35(137), 46–48.

Marek, G. R. (1982). Toscanini's memory. In U. Neisser (Ed.), Memory Observed: Remembering in Natural Contexts (pp. 508–511). San Francisco: Freeman.

Marcus, A. (1979). Great Pianists Speak. Neptune City, NJ: Paganiniana Publications.

Matthay, T. (1926). On Memorizing and Playing from Memory and on the Laws of Practice Generally. Oxford: Oxford University Press.

Monahan, B. J. (1978). The Art of Singing: A Compendium of Thoughts on Singing Published between 1877 and 1927. London: Scarecrow Press.

Miller, G. A. (1956). The magic number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81–93.

Nuki, M. (1984). Memorization of piano music. Psychologia, 27, 157–163.

Pauk, W. (1994). How to Study in College (5th ed.). Boston: Houghton-Mifflin.

(p.141) Persellin, D. C., & Pierce, C. (1988). Association of preference for modality to learning of rhythm patterns in music. Perceptual and Motor Skills, 67, 825–826.

Plunkett Greene, H. (1912). Interpretation in Song. London: Macmillan.

Price, G. E., & Dunn, R. (1997). Learning Style Inventory (LSI) Manual. Lawrence, KS: Price Systems.

Renzulli, J. S., & Dai, D. Y. (2001). Abilities, interests, and styles as aptitudes for learning: A person-situation interaction perspective. In R. J. Sternberg & L.-F. Zhang (Eds.), Perspectives on Thinking, Learning and Cognitive Styles (pp. 23–46). Mahwah, NJ: Erlbaum.

Revesz, G. (1925/1999). The Psychology of a Musical Prodigy. London: Routledge. (Original work published in 1925).

Ross, E. (1964). Improving facility in music memorization. Journal of Research in Music Education, 12, 269–278.

Rubin-Rabson, G. (1937). The Influence of Analytical Pre-Study in Memorizing Piano Music. New York: Archives of Psychology.

Sloboda, J. A., & Parker, D. H. (1985). Immediate recall of memories. In P. Howell, I. Cross, & R. West (Eds.), Musical Structure and Cognition (pp. 143–167). London: Academic Press.

Sloboda, J. A., Hermelin, B., & ƠConnor, N. (1985). An exceptional musical memory. Music Perception, 3, 155–170.

Small, G. (2003). The Memory Bible: Ten Commandments for Keeping your Brain Young. New York: Hyperion Books.

Sternberg, R. J., & Grigorenko, E. L. (2001). A capsule history of theory and research on styles. In R. J. Sternberg & L.-F. Zhang (Eds.), Perspectives on Thinking, Learning and Cognitive Styles (pp. 1–21). Mahwah, NJ: Erlbaum.

Wilding, J., & Valentine, E. (1997). Superior Memory. Hove, UK: Psychology Press.

Williamon, A. (1999). The value of performing from memory. Psychology of Music, 27, 84–95.

Williamon, A., & Valentine, E. (2002). The role of retrieval structures in memorizing music. Cognitive Psychology, 44, 1–32.

Williamon, A., Valentine, E., & Valentine, J. (2002). Shifting the focus of attention between levels of musical structure. European Journal of Cognitive Psychology, 14, 493–520. (p.142)