The Mind's Ear: I Hear Music and No One Is Performing

October 3, 2005

"He is a good musician, who understands the music without the score, and the score without the music. The ear should not need the eye, the eye should not need the (outward) ear."1

Hearing music in the mind's ear, without any sound source present, is a skill long prized by musicians. It is also the subject of books and articles by music educators and an area for recent research by psychologists and medical scientists. An early writer on the subject was Marie Agnew, who called the skill "auditory imagery," "tonal [pitch] imagery," and "inner hearing."2 One of the most prolific music educators on the subject, Edwin Gordon, calls it "audiation" and defines the concept broadly.3 For Gordon, audiation "takes place when one hears and comprehends music for which the sound is no longer or may never have been physically present." Gordon's definition goes beyond the skill and includes musical information about the tonal/harmonic/rhythmic structure of what is being imaged. Discussions of this skill are also present in sight-singing texts, such as that by Earl Henry, who shows the influence of Gordon in his definition of inner hearing as being "the internal mental processing of musical relationships."4 Andrea Halpern and Robert Zatorre, both prominent psychologists who have worked in this area, have studied the skill of audiation in trained musicians as well as in non-musicians.5


Why Is Hearing in the Mind's Ear Valued?

The mental hearing of music is an important tool for composers and performers. It frees composers from the constraints that may affect music composed at a keyboard and allows performers to rehearse nuances of style, unimpeded by the instrument. Henry Cowell described the importance of the inner ear in his compositional process. He said, "The most perfect instrument in the world is the composer's mind. Every conceivable tone-quality and beauty of nuance, every harmony and disharmony, or any number of simultaneous melodies can be heard at will by the trained composer."6 Robert Schumann encouraged other composers to use their inner ear ("do it [composition] all with your brain") rather than depend on a keyboard instrument.7 Carl Maria von Weber compared working with his internal hearing versus working at an instrument: "How differently does he work whose inner ear at once discovers and criticizes."8 Weber further describes how the inner ear comprehends tonal and formal structures, a concept that is echoed in Gordon's broader definition of audiation.

Noted performers have used inner hearing as an integral part of their practicing. Glenn Gould depended on mental practice, as did Vladimir Horowitz and Anton Rubenstein. A well-known piano teacher, Theodore Leschetizky, preached the value of inner hearing, as discussed by a pupil, Ethel Newcomb. She described how he would go on a walk to work out phrasing, tempos, and shading. She said that for him, "Listening to the inward singing of a phrase was of far more value than playing it a dozen times."9 Mental rehearsal is advocated by teachers and researchers. Malva Freymuth, a violin pedagogue, has written a book on the importance of mental practice and the development of the inner ear. She believes musicians should alternate continually between mental and physical practicing in order to develop "a strong musical memory."10

Those who are skilled in analysis know that if an analyst can use mental hearing, it is likely that relationships and deep structural understanding of the piece will emerge more clearly. In their leading Schenkerian analysis textbook, Allen Cadwallader and David Gagn encourage students to memorize a piece to be analyzed and to hear it "in one's inner ear."11 Carl Schachter and Felix Salzer, in their book Counterpoint in Composition, assume that students who use their text are able to hear the interaction of musical lines in two- and three-part textures. Another analyst, Steve Larson, discusses how the ability to hear sound internally is basic for hearing tension and resolution. He describes the need for hearing an implied resolution as primary for hearing instability in pitches.12


How Is this Skill Acquired?

Performers, composers and analysts acknowledge that inner hearing of music is an important skill. Although some people, such as Mozart,13 might seem to be born with a well-developed skill of mental hearing, most musicians describe their ability at inner hearing with varying levels of success and describe how they have to work at attaining it. Henry Cowell described in detail how he worked over a period of time to develop his ability to hear first melody and then harmony, and that he "had to make great effort to hear a violin tone,"14 implying that hearing timbre and harmony are more difficult than hearing melody.

Emile Jaques-Dalcroze believed that his combination of eurhythmics, singing, and improvisation could turn the whole person into an "internal ear."15 He believed that eurhythmics facilitated the "mutual integration of the individual's motor, perceptive, intellectual, and imaginative functions."16 Most eurhythmics training depends on musicians' reactions to actual sound rather than internalized sound, but the combination of aural and kinesthetic experiences does result in an enhanced conceptual understanding of many aspects of music. His view of internalized sound being related to motor, perceptual, and intellectual functions is supported by recent scientific research, as will be explained in a later section.

If inner hearing is valued by professional musicians and if it needs to be developed, the logical place to find it in the musical curriculum would be in performance studies and in aural training and sight singing courses. Indeed, aural training and sight-singing textbooks discuss the importance of inner hearing. Bruce Benward and Maureen Carr stress inner hearing's importance in the introduction to Sightsinging Complete: "The task of observing a musical score with thoughtful and hearing eyes is the most significant outcome of the four-semester sequence of courses" [first two years of music major study].17 Earl Henry in his text, Sight Singing, agrees with the importance placed on inner hearing. He affirms the coupling of inner hearing and sight singing, and believes that together they enable the development of musicianship more than any other kind of technique. Gary Karpinski calls the process of mentally hearing music "auralizing." He discusses the use of inner hearing (especially in intonation) and also relates it to sight-singing, but does not discuss specific ways to train inner hearing.18 Michael Rogers, in his book about music theory pedagogy, discusses inner hearing as "internal musical perception" and "aural imagery." He further couples it with singing and emphasizes the need to combine silent singing with regular singing in aural training teaching.19

Despite these strong affirmations for internal hearing and the coupling of inner hearing and singing, there is little instruction in sight singing texts for how to actually develop the skill; it is assumed that the skill will be a by-product of learning a number of pieces and developing the ability to sight-read over several years of applied lessons and sight-singing classes. A text that does address inner hearing is Sound Thinking, by Tacka and Houlahan, which is oriented to K-12 education and based on the Kodaly approach. This text, as can be expected, gives emphasis to singing but the curriculum goes far beyond being a sight-singing program with its inclusion of notation, memory, improvisation, and form and analysis. The writers believe that if a singing approach is used for training the ear, then a musician "will be able to hear music without needing to have it sounded aloud."20 Techniques for developing the inner hearing skill include recalling familiar songs, hearing familiar songs internally while tapping the beat or the rhythm, singing silently what is signed by the instructor (using Kodaly/Curwen hand signs), singing back familiar motives and phrases, and memorizing phrases and two-part examples from notation without hearing them aloud.

Collegiate aural training textbooks contain little instruction in how to develop the skill of mental hearing. Edward Klonoski has written several articles on the importance of auditory imagery and its lack of presence in aural skills courses. He does discuss how two current textbooks address the skill, at least indirectly. These texts are Functional Hearing and Listen and Sing.21

Another aural training textbook that includes inner hearing exercises is Ear Training and Sight Singing by Allen Trubitt and Robert Hines.22 Trubitt and Hines not only discuss the importance of inner hearing for good intonation, error detection, and sight-reading, but they also include a number of exercises intended to train inner hearing. One type of exercise, called "intonation," asks students to carefully and slowly sing intervals or pitch patterns against a reference pitch. This type of exercise encourages the increased acuity in detecting small pitch deviations. Other exercises, called "scanning," "anticipation," and "recognition of intervals," encourage students to look ahead and hear in their mind's ear the following measure. The tapes that accompany the text leave a measure of silence before the following interval or rhythm pattern is heard; students are to image the sound and then hear the performed sound. Another exercise, called "fusing," develops vertical listening. Students hear one pitch, imagine another, and then attempt to hear (or fuse) the two pitches together. Intervallic fusing is followed by harmonic fusing and the fusing of two-part music. Many ear-training and sight-singing texts that address the importance of inner hearing assume that its development will occur as a by-product of a traditional aural training curriculum. Trubitt and Hines, however, have not made that assumption and have deliberately included activities that train inner hearing in an intentional way.


Understanding of internal hearing from scientific research

Auditory imagery has been studied scientifically by psychologists in traditional behavioral studies and by psychologists and neuroscientists using brain lesions or scanning. When people have suffered a brain lesion through surgery or injury, scientists are able to examine what skills and knowledge have been lost because of the affected brain areas and are able to make judgments about how different areas function. For scientists choosing to use a scanning method, there are different technologies possible, including positron emission tomography (PET), electroencephalogram technology (EEG), and functional magnetic resonance imaging (fMRI). PET allows particular tasks to be related to specific brain structures with its excellent spatial resolution, but involves the injection of a low dosage of radioactive material. Functional MRI has the advantage of spatial resolution, but lacks good temporal information and is expensive to use. EEG scanning offers good temporal resolution but lacks the spatial resolution of PET or MRI; it is not a true three-dimensional scanning method. EEG scans are being used to study the timing of reactions to events with event related potentials (ERPs); for example, the mismatch negativity, which usually occurs around 120-200 milliseconds after the event, is often associated with a deviation from what was expected, such as an unexpected pitch. All of these techniques have been used in recent years to study music perception, auditory imagery, musical performance, emotion in music, and other topics. In the following sections I will discuss selected research that pertains to auditory imagery and what that research tells us about its use in performance and musical study.


Musical Performance and Auditory Imagery

Several researchers have examined and compared the brain activation patterns of musicians as they performed and then imagined performing a piece. One such researcher is Paula S. Washington who used EEG to study four string players performing violin-viola duos.23 The performers were asked to play a specific duo, to imagine playing the duo with and without the score, and to mentally hear one part while listening to the partner playing the other part. She discovered that the tasks in physical and mental performing were strongly correlated in brain activation patterns. That is, the brain areas used for each task were very similar and the tasks evidently recruit some of the same neural networks. Another researcher is Alvaro Pascual-Leone who studied where brain activation occurred as his subjects played and then mentally rehearsed a piano exercise.24 By using transcranial magnetic stimulation, he found that some of the same neural structures—namely prefrontal, supplementary motor area (SMA), basal ganglia and the cerebellum—were activated for both performing and imaging tasks. Mental rehearsal is thus activating the same motor areas that are activated when one is physically practicing. Pascual-Leone also observed a marked improvement in performance after subjects had mentally rehearsed the exercise. He believes the combination of mental rehearsal and physical practice should be used by musicians because it leads to greater improvement than using physical practice alone. These studies affirm the importance of mental rehearsal and provide a mandate for educators to train students in using internalization of sound.

Justin Sergent et al. studied pianists listening to music, then playing scales, and then reading a musical score.25 Using both PET and fMRI technologies, the researchers found that listening and reading activated all four lobes, including the occipital (visual area) and temporal (location of the auditory cortex) lobes of both hemispheres, as well as the inferior parietal lobe bilaterally and the left inferior frontal gyrus. The involvement of the parietal cortex suggests a mapping between notation and sound in an area adjacent to, but distinct from, the area where there is language mapping of words aurally and visually. When subjects combined listening, reading and playing, the superior parietal lobe was also used bilaterally. According to Sergent et al., this particular recruitment may reflect the observation that spatial information relative to notes on a staff is generated and is coordinated with the actual physical execution in performing music.

The final study I will discuss regarding performance concerned instrumental conductors with collegiate training. David Hoffman used EEG to study subjects as they read through three wind ensemble scores in their mind's ear.26 He found that there were differences in the degree of auditory imagery relative to a subject's familiarity with each score. This was especially apparent in the comparison of a subject's imaging a learned score versus an unfamiliar score. Hoffman also found that effective imagery recruited different cortical areas in the brain, some of them at considerable distance from each other. This was more pronounced for the learned scores. These four representative studies tell us that mental rehearsal is a complex task using widespread and multiple cortical locations. The implications are that developing skill in auditory imagery will require training and practicing by students and integrative tasks presented in multiple formats by instructors.


Internalization of Melodies and Chords

A number of studies use listeners' acquired tonal information as the basis for detecting the presence and activation patterns of internalized sound. Andrea Halpern, a leading scholar in auditory imagery, initially studied the topic through behavioral research. In one study, she used familiar tunes, such as "Row, Row, Row Your Boat" and "Yankee Doodle," that had likely been heard in only one key.27 She asked subjects to imagine the song and hum the starting pitch or find the starting pitch on the piano. There was consistency of pitches sung or played by subjects, indicating excellent long-term memories for keys. Daniel Levitin conducted a similar study and asked subjects to recall a favorite recording of popular music and sing the first note.28 Two examples were used: "Hotel California" by the Eagles and Madonna's "Get into the Groove." Forty percent of subjects were exactly correct on at least one trial and fifty per cent of subjects were within a half step. Like Halpern's research, this study also showed the accuracy of long term memory for pitches and keys.

In another study to explore memory for melody, Halpern examined melodic contour by giving subjects a song title and the lyrics and asking them whether the second word was higher than the first.29 The tunes that she selected included "Do Re Mi" and "The Star Spangled Banner." The subjects were successful with this task, although it was somewhat easier for musicians than non-musicians. Halpern and Robert Zatorre repeated this study using positron emission tomography (PET) to see which areas of the brain were activated during the task.30 The subjects' reaction times were related to the distance between words, indicating that songs were retrieved from memory and heard in the mind's ear. The task did not recruit the primary auditory cortex, but used areas adjacent to the auditory cortex and the SMA, as well as the thalamus and inferior frontopolar area, i.e., areas that are known to be involved in memory. The primary auditory cortex, located on the superior temporal gyrus or STG, is used for the perception of pitch, loudness, and timing. The adjacent secondary auditory area and the auditory association area are used for more complex aural tasks, such as the perception of timbre and melodies, and their tonal and metric organization. These findings show the role of memory, the complexity of auditory imagery, and the widespread neural networks involved in the task, as well as the strong similarities between music perception and music imaging.

Some tasks are more complex and take more time to perform. In a behavioral study, Timothy Hubbard and Keiko Stoeckig requested that subjects listen to a tone and then image a tone that would be a step higher.31 Subjects were also asked to hear a chord and then imagine a chord a step higher. The researchers closely observed response times and image-formation times. The internalization tasks were comparable to perception tasks in accuracy and timing, indicating that there are shared areas in the brain for perception and imagery tasks. The chord images took longer to generate than tone images, suggesting that more complex sounds are more difficult to imagine in the mind's ear.


Other Aspects of Auditory Imagery

The supplementary motor area (SMA) is typically activated in tasks of auditory imagery. There are two possible explanations for this activation. Subjects who play an instrument may be recruiting neural networks related to kinesthetic experiences with the instrument. Another probable explanation is that subjects are using subvocalization, the subject of a study by David Smith, Margaret Wilson, and Daniel Reisberg.32 These researchers looked at verbal and musical tasks that would likely use mental rehearsal and used techniques (such as chewing candy) to reduce the effectiveness of subvocalization for performing the experimental tasks. The musical task was to decide if a familiar melody rose or descended from the second note to the third note. These researchers found that in most (if not all) auditory imagery tasks, the inner voice plays an important role, and if it is blocked (as was done in their experiment), there is a significant decrease in performance. As there is likely an important link between subvocalization ability and sight-singing skills, this research offers scientific substantiation for what many aural training instructors have observed in years of teaching.

An important aspect of auditory imagery of familiar music is the retrieval of music from memory. Halpern and Zatorre used PET to study brain activation when subjects listened to the first few notes of a familiar tune and were asked to continue hearing that tune in their minds' ears.33 They used instrumental tunes without text to minimize the effect of recruiting verbal processing networks in the left hemisphere. The hypothesis that the right hemisphere would be emphasized when the melodies did not have text was proved, and led the authors to confirm that the right hemisphere is specialized for processing tonal imagery. The activation of brain areas such as the right inferior frontal, bilateral frontal, and the right temporal, showed that imagery involved retrieval from semantic memory. One interpretation might be that episodic memory is used to recall familiar melodies, and according to the researchers, this could have occurred, but the major retrieval was semantic (conceptual) rather than episodic (specific event).

The unintended hearing of music in the mind's ear is a sub-category of hallucinogenic sounds and voices. Musical hallucinations are caused by a hearing impairment, brain injury, epilepsy, intoxication, or psychiatric problems.34 Women and people older than 60 are somewhat more prone to suffer from these types of hallucinations, and some composers, such as Smetana and Ravel (who suffered from neurodegenerative disease in later life), reported being flooded with music in their heads.35 Haydn was another composer who in later years complained of being overwhelmed with musical ideas. According to an early biographer, in conversation during the year 1806 Haydn said, "Usually musical ideas are pursuing me, to the point of torture. I cannot escape them, they stand like walls before me. If it's an allegro that pursues me, my pulse keeps beating faster, I can get no sleep. If it's an adagio, then I notice my pulse beating slowly. My imagination plays on me as if I were a clavier."36


Summary of auditory imagery research

Scientific research in auditory imagery is still in its infancy. There are many questions about how musicians process music, how musical information is stored and retrieved, how widespread neural circuits are integrated for music processing, and how musicians can develop and refine their ability in auditory imagery. There are, however, aspects about auditory imagery that we know and that affect pedagogical approaches and strategies for developing this important skill.

  • Auditory imagery is an acquired skill. Most people, trained musicians and untrained listeners, experience auditory imagery, but the brain activation patterns are different for the two kinds of listeners, and training facilitates the speed and complexity of what is heard in the mind's ear.
  • Auditory imagery is strongly correlated with aural perception, in that similar and often the same brain areas are activated. The implications for training the skill of auditory imagery are that similar techniques may be used, but there is the need to find additional ways of training and refining the skill.
  • Auditory imagery is not highly dependent on the primary auditory region but it is the associative areas that handle more complicated music processing. Thus the skill is a complex one, integrating a number of neural networks. In auditory imagery, all four lobes and both hemispheres are used. This fact and the demonstrated communication between distant areas of the brain likewise support the characteristic of complexity. For pedagogy, then, the training of this skill demands a variety of tasks that are integrated with different ways of relating to music and to multiple aspects of music.
  • Research supports the assumption that learned scores are much easier to hear in the mind's ear. In the use of familiar scores, a musician is recalling already established musical patterns.
  • Harmony takes longer and is more difficult to image than single-line melodies. Musicians might assume this, but it has also been demonstrated by researchers.
  • Pascual-Leone and others showed the importance of mental rehearsal to accompany physical rehearsal for a musician. It is important to remind students of the need for mental rehearsal and to train them as well in the skill of imaging unknown scores.
  • In auditory imagery, the supplemental motor area (SMA) is recruited in a variety of tasks. This recruitment is likely related to subvocalization and probably related to a person's instrumental study as well. The implications for pedagogy are seen in the importance of relating imaging tasks to the students' instruments, and the importance of sight-singing instruction in the aural skills program.


Understanding Perception and Mental Hearing as Schema

Schematic Representations

The complexity of hearing, as demonstrated by neuronal activation, is demonstrated also in a more psychological understanding of schematic representation. Schema are mental representations of related entities. Schema do, however, have a neural basis, even as psychological representations. In the words of Bergan, for example, the perception of pitch requires "the utilization of mental processes that rely on a mental representation or tonal image of the sound itself, that is a tonal image that closely approximates the actual experience of hearing against which one compares actual pitch."37 Schema are constantly changing and adapting through our experiences, and are thus characterized as "dynamic." The adaptability and changeability of schema represent neuronal change and adaptability in the brain (which is called plasticity). A musical task has multiple schema that are inter-related. In musical performance, the schema that relate to playing an instrument are actually multiple representations that change with each listening, practicing, and performing experience. The network of schema and their overlay represent well the complexity of music cognition.


Schema in Music Listening

Schema are at work when we listen to music. If the piece is a very familiar one, we use a veridical memory (for the specific piece) and have specific expectations throughout the listening. If the piece is not as familiar or is unknown to us, schema for metric and pitch structures are at work in setting forth expectations of what will occur next. These expectations of what to anticipate are based on past experiences and help us to organize the music in a dynamic way during the listening process. Our expectations affect how we will interpret consonance and dissonance, tension and repose.


Schema in Mental Hearing

A schematic representation is helpful for understanding the nature of music imaging. While we may have an actual or veridical representation of music pieces that we know very well (so that the whole piece could "play" in our mind's ear like a tape recording), most of our musical imaging does not work in that way. Rather than mentally storing complete pieces, we likely have representations or schema of familiar pitch or rhythmic patterns, schema of metric structures, and schema of tonal or harmonic structures. Some of these may be rather involved networks of schema (such as our understanding of tonality), while others may be small in musical terms (such as an interval or a simple rhythmic pattern).

If we mentally hear music from a score, with no sound present, then we activate schema that represent metric structures, rhythmic and melodic patterns, tonal structure, timbral qualities, etc. If the score is a familiar piece, then the schema are more complex and more highly developed. If the score is unfamiliar, then we likely are using lower level schema, as indicated by the fact that this kind of mental hearing does not happen with the same facility as imaging a familiar piece.

The act of mentally hearing a score with no sound present is a re-constructive or re-assembly process—we assemble bits and pieces of sounds that are already represented in memory and construct the sounds of the score before us. Lehmann and McArthur call it a "reconstructive activity."38 As an example of this re-assembly process, consider how one might image a chord in SATB voicing. One could literally add together the intervals that comprise the chord. Or, one could recall the sound of the triad (whether major or minor), imagine an arpeggiation of the chord pitches, and then arrange the notes in the particular voicing used in the notated chord. There is, then, a re-assembly of recalled sounds. When musicians are termed "good sight readers," they likely possess facility in this re-assembly processing.


Given the foregoing evidence and conclusions, the ability to hear music in one's inner ear is indeed a very complex task. It is dependent on what is in memory, on our experiences in listening, studying and performing, and on our flexibility and facility in using our musical schema or representations. This complex skill is deemed a necessary skill for musicians, be they performers, composers, or critical listeners. How and when does the training for this complex skill occur? Altenmüller believes that musical learning should include the implanting and integration of mental images of sound, and that this kind of training is so basic for musicians that it ought to occur before written theory skills are taught. Is the skill being addressed in the studio, in ensemble rehearsals, and in aural training classes? Or are we hoping that the teaching of the technical and expressive aspects of performance, emphasizing good intonation and blended sound in ensembles, and drilling identification of intervals and sonorities along with melodic and harmonic dictation will somehow train and enable students to acquire the complex skill of imaging music with acuity?


Techniques for developing internal hearing

This section of the paper will present some ideas and techniques that can be used by individuals who want to develop further their own mental hearing, and by studio and aural skills teachers who want to encourage their students in the development of this important skill. Most of these techniques are ones that I have used in the classroom, in working with students, or in my own practice and study of music. Increasing one's ability to use inner hearing takes practice and concentration, and so there should be purposeful time for trying these methods. If you teach, do not just tell a student to practice a particular activity; rather, lead the student through a simple exercisegive the space and time in class or lesson for using inner hearing. We cannot assume that it will be simply a by-product of aural/kinesthetic practicing.


Hearing melodies in the mind's ear:

In order to develop an ability to internalize sound, use the memory that we have for melodies. If a melody is familiar, such as "Row, Row, Row Your Boat," one can simply recall it and hear it in the mind's ear. If the melody is unfamiliar, a technique is to hear it one phrase at a time and memorize each phrase. For example, play a phrase on an instrument, pause and intentionally listen to it in the mind's ear, and then sing it. Do this with several phrases, and then listen to the group of phrases in the mind's ear.

Extend the mental hearing of familiar and unfamiliar melodies by adding features to the mind's image. For example, ask students to imagine "Row, Row, Row Your Boat" with dynamic shaping or with different kinds of articulation, or played on different instruments. Two further extensions would be to hear the melody in another mode or to take a motive and mentally hear it in a tonal sequence or a real sequence. Another extension is to identify hyper-measures and to hear (mentally) the strong beat emphasis at the measure and at the hyper-measure levels. Two familiar melodies to use are "O Danny Boy" and "I Could Have Danced All Night" (from My Fair Lady).

Another idea for the mental hearing of familiar melodies is error detection. In aural training classes, the instructor can provide the notation of a familiar tune but with pitch and rhythmal errors. The incorrectly notated melody should not be played or sung, so that students are relying solely on their mind's ear to detect the errors.

For pieces being studied in applied lessons, students can refine the dynamic and timbral shape of each phrase as they listen in their mind's ear. They should be encouraged to try different ways of playing a phrase in their mind's ear and copying those ways with their instrument. Students should be trained to use their mind's ear to direct their playing rather than using the ear to react to sound.


Relate melodies to an instrument:

When students are hearing a melody in their mind's ear, encourage them either to finger a familiar instrument (touching the keys without actually playing) or to imagine fingering an instrument. This technique is particularly helpful for students in aural skills classes when they are engaged in sight-singing or dictation tasks. Since scientific research has demonstrated the involvement of the supplemental motor area in auditory imagery, these neural connections should be used and strengthened.


Relate melodic phrases to harmony:

For sight-singing assignments, I encourage students to decide what chords might accompany the melodies; these may be indicated with either Roman numerals or pop symbols. Then students are to accompany themselves with chords as they sing. They may use simple, closed position triads played with one hand, since the intention of the exercise is the integration of melody and harmony, not the voice leading of the accompanying chords. They may wish to practice this way phrase by phrase. After practicing several times with a keyboard (or guitar), they should sing accompanied only with the sounds of the chords in their mind's ear. The students' intonation will indicate which chords they are able to hear and which they cannot yet hear in their mind's ear.

For performance pieces, I would use a similar methodology, but encourage the use of pop symbols, especially for more recent compositions. This methodology encourages the mental integration of melody and harmony and makes students more aware of the accompaniment and how their part fits with it. As shown by research, harmonic listening is more difficult and demands purposeful practice. It is very important for performers to develop and refine the ability to use mental rehearsal as a practicing technique. The scientific research has shown that for a semi-professional or professional performer, this method is just as effective and complements kinesthetic rehearsal. Anecdotal evidence that I have collected from mature performers is very supportive of the research in the efficacy of mental rehearsal.


Converting internalized sound into notation:

Dictation and notation of familiar and unfamiliar melodies are common tasks in aural training classrooms. Advanced students may assume transcription projects of jazz solos or concerto cadenzas. For a composer, the refined ability to internalize melody, harmony, timbral distinctions, and subtle dynamic and articulation changes is critical. Students need to be challenged with a variety of tasks to convert internalized sound into notation. For each task, some level of memorization and knowledge of the tonal and harmonic structure of the style are needed. There are numerous sources that offer suggestions for training melodic and harmonic dictation. See Michael Rogers' Teaching Approaches in Music Theory, second edition, and his bibliography for additional sources.39


Turning internalized sound into performance:

Musicians who "play by ear" with facility have been envied by those who are not able to do it easily. It is an ability that can be developed and students should be encouraged to do so. In recent years, improvisation, which has always been prominent in jazz studies, has achieved a more important place in other parts of music curricula, such as in aural training classes. There are numerous sources on improvisation (classical, popular and jazz) and improvisational exercises are discussed in recent aural skills textbooks.40


Mental hearing from notation of more than one line:

Listen to recordings of pieces and closely follow the score, not only the solo part but the accompaniment as well. Try to follow more than one line at a time. Although this technique seems fundamental, a surprising number of students (even graduate students) have confessed that they listen to recordings with a score but follow only their own part.

Use the mind's ear to hear harmonic intervals. If they are hard to hear, then internalize them as two melodic notes and gradually bring the notes together in the mind's ear until they are heard simultaneously. Then extend the exercise by transposing one of the pitches an octave up or down so that simple and compound intervals can be heard easily. Another extension is to transpose intervals up or down so that they do not necessarily occupy one's vocal range. A student needs to be comfortable with harmonic intervals in different pitch spaces before hearing harmonies.

To listen to a two-line piece, such as species counterpoint or two parts of a choral piece, apply the skills of hearing intervals with the mind's ear. One may need to proceed in a zig-zag fashion, hearing the notes of each interval in a melodic way and bringing them together as a simultaneity. It is also helpful to hear each line, perhaps a phrase at a time, as its own melody in the mind's ear. As with melodies, this skill can be extended by adding articulation and dynamics. One can also add timbre, with the same or different timbres on each line.


Mental hearing of harmony:

In order to hear harmony in the mind's ear, begin with three- and four-note tertian chords in root position and in inversions. Start with the lowest or highest pitch and mentally hear the intervals of the chord melodically. Then slowly bring the pitches together until they are heard as a simultaneity. As with intervals, extend this exercise by transposing the lowest note down an octave, or trying different spacings for each chord, all in the mind's ear.

To begin hearing a chord progression, one might start with a simple succession of intervals in notation or in sight-singing shorthand. For example, one could hear mentally the diatonic pitches 1-3-5, 4-6-8, 5-7-9, 8-5-3-1. Pause after each chord and hear the pitches as a vertical arrangement, then mentally hear the chord outlines as a series of chords. This is a good beginning exercise for those who play a "single line" instrument, for whom hearing harmony is more difficult.

Another harmonic technique is to use a familiar melody such as "Amazing Grace." Ask students to play the melody by ear on a keyboard and then add harmony to it, again by ear. You might also try giving students the text of this melody and have them decide the harmony using only their mind's ear, away from a keyboard.

The above exercises are preliminaries for mentally hearing notated harmony. To work with notated harmony, begin with a simple hymn. For each chord, hear the notes of the chord as a compilation of melodic intervals that are then brought together in the mind's ear. As one moves from chord to chord, the mind's ear also needs to listen to the voice leading connections for each part. At first, this seems laborious and slow, but it gets easier and faster with practice. As the students' skills develop, work with a Bach chorale, which has more movement and more non-chord tones.

Similar strategies can be applied to mentally hearing a piano sonata by Mozart or a symphonic movement by Haydn or a Lied by Schubert. An additional strategy that can be used by an instrumentalist or vocalist studying an applied piece is to hear each line separately, i.e., the solo part, the right hand of the accompaniment, and the left-hand part. When this is comfortable, mentally hear combinations of two parts, and then put all three together. It may be helpful to work in phrase units.


Concluding Comments

In this article I have emphasized the importance of mental hearing for all performers, composers, and critical listeners. Teachers often expect their students to already possess this skill, or they expect that it will emerge as a by-product of practicing an instrument, or of observant performing in ensemble settings, taking sight-singing and aural-training classes, or listening carefully to performances and recordings. There is, therefore, little formal or even informal attention to how this skill may be taught; students are left to their own devices. Because of my interest in mental hearing, I have discussed the topic with professional performers. Allow me to share three things that they voiced. First, there seems to be a strong connection between mental hearing and some sort of kinesthetic awareness, related to what the professional does now or did as a child and teenager. This kinesthetic connection might be in the form of conducting gestures, trombone positions, clarinet fingerings, touching the strings of a cello, or playing the piano. There were, therefore, strong links to a practitioner's experience, affirming at least for these musicians a likely link to the supplementary motor area of the brain during mental hearing. A second observation was the importance placed on singing as an indication of how a musician might shape a phrase or a person's sense of intonation. This link to singing was stressed by instrumental performers, not only choral conductors and vocalists. One performer said that if a student cannot sing a line, then that person cannot play the line. The final observation, which validates the importance of mental hearing for a musician, is that performers consistently spoke of mental hearing as preceding what emerged from an instrument. That is, mental hearing is recognized as being more accurate than performing and needs to direct the actual performance. Working out phrasing and fine-tuning one's acuity for pitch need to occur in one's mental ear. The inner ear provides the leadership for performing—mental hearing is the leader for the next note, the dynamic shape of a passage, and intonation. One performer called the skill "hearing in the future." When we enable the development of a student's mental hearing we are enhancing that person's future in music, whether one note or one phrase or whatever time period ahead.


Agnew, Marie. "The Auditory Imagery of Great Composers." Psychological Monographs 31 (1922): 279-87.

. "A Comparison of the Auditory Images of Musicians, Psychologists and Children." Psychological Monographs 31 (1922): 268-78.

Alonso, Robert J., and Robert M. Pascuzzi. "Ravel's Neurological Illness." Seminars in Neurology 19 (1999): 53-57.

Altenmüller, Eckart, and Wilfried Gruhn. Music, the Brain, and Music Learning. Narberth, PA: Gordon Institute for Music Learning, 1997.

Bachmann, Marie-Laure. Dalcroze Today: An Education Through and into Music. Oxford: Clarendon Press, 1991.

Benward, Bruce, and Maureen Carr. Sightsinging Complete. New York: McGraw-Hill, 1999.

Bergan, John. "Pitch Perception, Imagery, and Regression in the Service of the Ego." Journal of Research in Music Education 13 (1965): 15-32.

Boursy, Richard. "The Mystique of the Sistine Chapel Choir in the Romantic Era." Journal of Musicology 11, no. 3 (Summer 1993): 282-83.

Cadwallader, Allen, and David Gagné. Analysis of Tonal Music: A Schenkerian Approach. New York: Oxford University Press, 1998.

Cowell, Henry. "The Process of Musical Creation." American Journal of Psychology 37 (1926): 234-35.

Damschroder, David. Listen and Sing. New York: Schirmer Books, 1995.

Dies, Albert C. Joseph Haydn, trans. Vernon Gotwals. Madison: University of Wisconsin Press, 1963.

Evers, Stefan, and Tanja Ellger. "The Clinical Spectrum of Musical Hallucinations." Journal of the Neurological Sciences 227 (December 2004): 55-65.

Freymuth, Malva. Mental Practice and Imagery for Musicians. Boulder: Integrated Musicians Press, 1999.

Gordon, Edwin. Learning Sequences in Music. Chicago: G.I.A. Publications, 1980.

Gottschalk, Arthur, and Phillip Kloeckner. Functional Hearing. New York: Ardsley House Publishers, 1997.

Halpern, Andrea. "Memory for the Absolute Pitch of Familiar Songs." Memory and Cognition 17 (September 1989): 572-81.

. "Mental Scanning in Auditory Imagery for Songs." Journal of Experimental Psychology: Learning, Memory and Cognition 14 (July 1988): 434-43.

Halpern, Andrea, and Robert Zatorre. "When That Tune Runs Through Your Head: A PET Investigation of Auditory Imagery for Familiar Melodies." Cerebral Cortex 9 (November 1999): 697-704.

Henry, Earl. Sight Singing. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1997.

Hoffman, David K. "Auditory Imagery of Conductors: an Examination of the Electroencephalographic Correlates of Score Reading before and after Score Study." Ph.D. dissertation, University of Minnesota, 2002.

Hubbard, Timothy, and Keiko Stoeckig. "Musical Imagery: Generation of Tones and Chords." Journal of Experimental Psychology: Learning, Memory and Cognition 14 (October 1988): 656-67.

Karpinski, Gary. Aural Skills Acquisition: the Development of Listening, Reading, and Performing Skills in College-Level Musicians. New York: Oxford University Press, 2000.

Klonoski, Edward. "Teaching Pitch Internalization Processes." Journal of Music Theory Pedagogy 12 (1998): 81-96.

Klonoski, Edward and E.A. Johnson. "Connecting the Inner Ear and the Voice." Choral Journal 44 (October 2003): 35-40.

Larson, Steve. "Scale-Degree Function: A Theory of Expressive Meaning and its Application to Aural Skills Pedagogy." Journal of Music Theory Pedagogy 7 (1993): 69-84.

Lehmann, Andreas, and Victoria McArthur. "Sight-Reading." In The Science and Psychology of Music Performance, ed. Richard Parncutt, 135-50. New York: Oxford University Press, 2002.

Levitin, Daniel J., and Perry R. Cook. "Memory for Musical Tempo." Perception and Psychophysics 58 (August 1996): 927-35.

Mursell, James. The Psychology of Music. New York: Norton, 1937.

Newcomb, Ethel. Leschetizky As I Knew Him. New York: Da Capo Press, 1967.

Pascual-Leone, Alvaro. "The Brain That Plays Music and Is Changed by It." Annals of the New York Academy of Sciences 930 (2001): 315-29.

Rogers, Michael. Teaching Approaches in Music Theory: An Overview of Pedagogical Philosophies. 2nd ed. Carbondale: Southern Illinois University Press, 2004.

Salzer, Felix, and Carl Schachter. Counterpoint in Composition. New York: McGraw-Hill, 1969.

Schumann, Robert. Music and Musicians. Essays and Criticisms, trans. Fanny Ritter. London: William Reeves, 1876.

Seashore, Carl. Psychology of Music. New York: McGraw-Hill Book Company, 1938.

Sergent, Justin, Eric Zuck, Sean Terriah, and Brennan MacDonald. "Distributed Neural Network Underlying Musical Sight-Reading and Keyboard Performance." Science 257 (July 1992): 106-9.

Smith, J. David, Margaret Wilson, and Daniel Reisberg. "The Role of Subvocalization in Auditory Imagery." Neuropsychologia 33 (November 1995): 1433-54.

Tacka, Philip, and Michael Houlahan. Sound Thinking: Developing Musical Literacy. New York: Boosey and Hawkes, 1995.

Trubitt, Allen, and Robert Hines. Ear Training and Sight Singing. New York: Schirmer, 1980.

Washington, Paula S. "An Electroencephalographic Study of Musical Performance: Imagined versus Actual Playing and Solo versus Chamber Playing." Ph.D. dissertation, New York University, 1994.

Zatorre, Robert, and Andrea Halpern, et al. "Hearing in the Mind's Ear: A PET Investigation of Musical Imagery and Perception." Journal of Cognitive Neuroscience 8 (1996): 29-46.

1Schumann, Music and Musicians, 63.

2There were two articles published by her: "A Comparison of the Auditory Images of Musicians, Psychologists and Children" and "The Auditory Imagery of Great Composers." These articles were published posthumously and there were no bibliographies. Agnew was a student of Carl Seashore's and both of her research studies are described in some detail, with references, in Carl Seashore, Psychology of Music.

3Gordon, Learning Sequences in Music, 3. The web site of the Gordon Institute for Learning lists publications and other information about Gordon:

4Henry, Sight Singing, xiv.

5See "References" at the end of this paper.

6Cowell, "The Process of Musical Creation," 234.

7Schumann, Music and Musicians, 417.

8Mursell, The Psychology of Music.

9Newcomb, Leschetizky As I Knew Him, 18-19.

10Freymuth, Mental Practice and Imagery for Musicians.

11Cadwallader and Gagné, Analysis of Tonal Music, vii.

12Larson, "Scale-Degree Function," 69-84.

13One is reminded of the story of how Mozart wrote out the choral piece Miserere by Allegri after the Sistine Chapel refused to give him a copy of the score. He wrote it after one listening and used a subsequent hearing to correct any errors (there were not many). There is an account of this in Richard Boursy, "The Mystique of the Sistine Chapel Choir in the Romantic Era," 282-83.

14Cowell, "The Process of Musical Creation," 235.

15Bachmann, Dalcroze Today, 40.

16Ibid., p. 298.

17Benward and Carr, Sightsinging Complete, xi.

18Karpinski, Aural Skills Acquisition, 153-56.

19Rogers, Teaching Approaches in Music Theory, 127.

20Tacka and Houlahan, Sound Thinking, 4.

21Gottschalk and Kloeckner, Functional Hearing; Damschroder, Listen and Sing.

22Trubitt and Hines, Ear Training and Sight Singing.

23Washington, "An Electroencephalographic Study of Musical Performance."

24Pascual-Leone, "The Brain That Plays Music and Is Changed by It."

25Sergent, et al., "Distributed Neural Network," 106-9.

26Hoffman, "Auditory Imagery of Conductors."

27Halpern, "Memory for the Absolute Pitch of Familiar Songs."

28Levitin and Cook, "Memory for Musical Tempo."

29Halpern, "Mental Scanning in Auditory Imagery for Songs."

30Zatorre and Halpern, et al., "Hearing in the Mind's Ear."

31Hubbard and Stoeckig, "Musical Imagery: Generation of Tones and Chords."

32Smith, et al., "The Role of Subvocalization in Auditory Imagery."

33Halpern and Zatorre, "When That Tune Runs Through Your Head."

34Evers and Ellger, "The Clinical Spectrum of Musical Hallucinations."

35Alonso and Pascuzzi, "Ravel's Neurological Illness."

36Dies, Joseph Haydn, 141.

37Bergan, "Pitch Perception, Imagery, and Regression."

38Lehmann and McArthur, "Sight-Reading."

39Rogers, Teaching Approaches in Music Theory.

40See for example this author's "Improvisation in the Aural Curriculum: An Imperative," College Music Symposium 37 (1997): 49-64. See also Joel Phillips, Jane Piper Clendinning, and Elizabeth West Marvin, The Musician's Guide to Aural Skills, (New York: W.W. Norton) 2005.

16444 Last modified on October 4, 2018