ARTICLE 63

Psychology of Music
Memorization by a jazz musician: Psychology oj Music
Copyright © 2008
a case study Society for Education. Music
and Psychology Research
vol 36(r): 63-79 [0305-7356
(20080I) 36:r; 63-79]
IO.II77/0305735607080834
http://pom.sagepub.com

HELGA NOICE, JOHN JEFFREY AND TONY NOICE

ELMHURST COLLEGE, USA
ROGER CHAFFIN
UNIVERSITY OF CONNECTICUT, USA

ABSTRACT To investigate the memory strategies of jazz musicians, we videotaped an

experienced jazz pianist as he learned a new bebop piece. He had not previously heard
a recording of the selection, nor had he seen the written music. The pianist provided
detailed reports of the musical structure and the types of cues he used as landmarks
to guide his memorization. Analysis of the videotapes, verbal reports, and multiple
annotated copies of the music revealed that this jazz pianist's learning process was
similar to that reported for classical musicians. That is, he used the musical structure
as a retrieval scheme and practiced using performance cues to elicit lmowledge of
upcoming passages from long-term memory. However, this study looked only at the
learning strategy for the note-by-note renditions that typically comprise the first
choruses of jazz performances. The relationship between this original memorization
process and the ability to improvise subsequent choruses will have to be addressed in
future investigations.

KEYWORDS: cognition, expertise, learning, music

How do jazz musicians learn new music? Unlike performers in the western art music
tradition in which the musician plays the score precisely as written, the jazz artist
varies the melody, rhythm, even the time-signature from one performance to the next
(Berliner, 1994; Sudnow, 2001). This quality of spontaneous variation lies at the
heart of the difference between classical music and jazz. In both traditions, the per-
former strives to produce an interpretation that is unique and personal. Classical per-
formers typically work out their interpretations in the privacy of the practice studio
and then reproduce essentially the same rendition from one public performance to
the next (Chaffin et aI., 2007; Shaffer, 1984; Shaffer et al., 1985; see Gabrielsson,
1999, for a review). Jazz musicians, in contrast, do not produce the same perform-
ance repeatedly. Instead, they rely on their feelings of the moment to spark novel
paths through each performance, bolstered, when needed or desired, by their 'bag of
tricks' (e.g. favorite chord inversions or chord voicings). At his or her best, a jazz musi-
cian functions as a second composer, frequently making so many spontaneous

sempre
64 Psychology of Music 36(1)

changes that the original creator might not recognize parts of his or her composition
(Ward, 2000).
How do jazz musicians memorize a new piece when each performance will be dif-
ferent? Is the process similar to preparing a new piece in the classical idiom, or is it
as different, as the very distinct performance conventions of the jazz and classical
idioms suggest? Relatively little attention has been paid to the question. Lehmann and
Gruber (2006) suggest that the stages for acquisition of improvisational ability are
the same as those demonstrated for composition, and other researchers (e.g. Berliner,
1994; Pressing, 2000; Sloboda, 1994) have investigated the various abilities involved
in playing jazz.
The differing performance expectations for the jazz and classical idioms suggest that
musicians might use different memorization strategies for the two types of music. Scores
for classical music are typically written out fully,containing every note to be played by
the right and left hand, and performers are expected to faithfully reproduce every note.
As a result, details of technique are typically settled and practiced so that motor mem-
ory is automatic. Most jazz selections, in contrast, are learned from lead sheets, which
provide only the single notes of the melody, with chord symbols sketched in above them.
Jazz musicians are expected to improvise, varying key, melody, and rhythm in each per-
formance in spontaneous invention (e.g., Lehmann, 2005). Creative decisions are made
in real time (Kenny and Gellrich, 2002), and the resulting melodies and harmonies fre-
quently surprise even the performer (Berliner, 1994).
On the other hand, there are also important similarities between the jazz and
classical idioms that might encourage the use of similar memorization strategies. In
both idioms, performers must express the musical feelings and ideas represented by
the melodic and harmonic structure of a piece. In jazz, this abstract conception of the
piece is implemented as a sequence of chord changes using flexible rules enhanced
with prelearned, integrated motor sequences that make up each artist's riffs, chops,
or the aforementioned bag of tricks (Sudnow, 2001; Ward, 2000). Similarly, per-
formers in the classical idiom have to understand the underlying structure of the
piece in order to play the notes that the composer has specified with suitable musical
expression.
Studies of musicians in the western art music tradition suggest that the abstract,
conceptual representation of the melodic and harmonic structure plays a central role
in their memorization. Hallam (199Sa, 1995b, 1997, 2001) found that half of the
experienced professional musicians that she interviewed used the hierarchical struc-
ture of the music to organize their memory for a new piece. The other half may also
have used the musical structure for memorization, but this could not be determined
from their self-reports; they reported memorizing automatically, without explicitly
using any deliberate strategy.
Direct observation of practice also indicates that musical structure plays a central role
in memorization for classical music. In a series of longitudinal case studies, Chaffm and
his associates have described the learning processes of a classical pianist, singer, cellist
and conductor (Chaffm, 2007; Chaffm and Imreh, 1997, 2001; Chaffin et al., 2002,
2003, 2007; Ginsborg et al., 2006; Imreh and Chaffin, 1996/97). They propose that
experienced classical performers memorize in much the same way as expert memorists
 Noice et al.: Memorization by a jazz musician 65

in other domains. Likeother expert memorists, the learning strategies of these musicians
can be described in terms of skilled memory theory (Chase and Ericsson, 1982) and
its subsequent refmement, long-term working memory (Ericsson and Kintsch, 1995).
Experienced musicians make use of three strategies employed by expert memorists in
many declarative memory tasks: chunking, organization and practice. Like skilled mem-
orizers in other domains, classical musicians organize the material they must remember
into chunks (e.g. grouping notes into familiar patterns such as scales and arpeggios), use
the organization provided by the musical structure as a hierarchical retrieval scheme to
organize memory for the music and engage in extended practice of retrieval from long-
term memory.
An important additional strategy used by the classical musicians studied by Chaffin
and colleagues (e.g. Chaffin et aI., 2002) was their selection of particular features of the
music to serve as retrieval cues to elicit memory for the piece as it unfolds during per-
formance. These performance cues serve as landmarks to guide performance. Attention to
performance cues provides the classical musician with cognitive control of the highly
practiced and automatic movements of the performance. Other studies of classical musi-
cians have also revealed the same use of the formal structure and extensive overlearn-
ing (e.g. Williamon and Valentine, 2002). Thus, starting to play at a particular location
establishes a link between the representation of the music in working memory and the
motor sequences needed to produce the notes. Thinking of the music becomes a retrieval
cue, automatically eliciting the highly practiced movements of the performance.
Extended practice is needed to make performance cues function rapidly and reliably
enough to perform under the pressures and distractions of a liveconcert. One study docu-
mented 10 hours of practice for each minute of performance (Chaffin and Imreh, 2002),
and Lehmarm and Ericsson (1998) studied an advanced student pianist who devoted
15 hours of practice for each minute of performance.
The encoding and retrieval processes of other types of performing artists have also
been shown to involve the same memory principles. Two of the authors of this article
have written extensively on the learning strategies of professional actors (for reviews,
see Noice and Noice, 2002, 2006) and have found that these strategies also appear con-
sistent with the principles of both slalled memory theory (Chase and Ericsson, 1982)
and long-term working memory (Ericsson and Kintsch, 1995). That is, the actor pos-
sesses domain-specific information in long-term memory on human motivations and
the myriad specific objectives that are used to act upon those motivations. The text input
is classified in terms of these objectives. ('In this section, I'm trying to soften him up.
Then, in the next section, I go in for the kill.') These objectives then serve as efficient
retrieval cues and the entire succession of objectives constitute a retrieval structure for
the complete role. Thus, residing in long-term memory are both the exact dialogue and
the necessary performance information (Le. the reasons for uttering the dialogue). As
already pointed out, this memory model would have equal application to musicians. The
formal structure of the music and the performance cues established during practice
would constitute a highly efficient retrieval structure.
However, the particular nature of the actor's strategy results in a script-acquisition
process that bears similarities to both classical and jazz music. That is, just as the clas-
sical musician renders the score with note-for-note fidelity, the actor must render the
66 Psychology of Music 36(1)

original script with word-far-word fidelity. However, the classical musician generally
plans the expressive cues in advance, deciding that certain passages will be played with
gaiety, surprise or excitement (Chaffin and Imreh, 2002). Of course, some changes
occur spontaneously in classical concerts, but they are more of degree than of kind
(e.g. varying the dynamic range of various passages), resulting in certain unique in-
effable musical feelings being communicated at every performance. This is frequently
true for acting, as well; however, on some occasions, actors actually change the under-
lying meaning of their utterances so that the same words may be defiant in one per-
formance and apologetic the next. In this sense, actors are more akin to jazz musicians.
That is, depending on their impulses at the moment of performance, today's scene of
confrontation may become tomorrow's scene of conciliation.
For example, two actors (playing husband and wife) might have the following
exchange: The male actor says, 'Look, don't argue, just do it.' The female actor replies,
'That's it? Just do it?' The husband's intention (derived from an analysis of the script) is
to malce an angry demand of his wife. Her intention (also derived from gleaning the
playwright's intended meaning) is to challenge her husband's right to make such a
demand. However, one night, the male actor might spontaneously use his dialogue more
as a confession of desperation than an attack. (Why it comes out that way, the actor
probably doesn't even know - but most professionals report that such spontaneous
changes do occur on stage from time to time.) Because the female actor is closely
attuned to the male actor's utterances, she responds 'in the moment' to this new mean-
ing with a slight smile. Her line of dialogue then becomes wryly conciliatory. It might
be thought that the actors' openness to such impulses could distort the play's meaning.
However, well-written plays can almost always accommodate spontaneous acting with-
out violating the playwright's intentions. The characters of Blanche and Stanley in
A Streetcar Named Desire (Williams, 1947) are so well delineated that the play's overall
thrust will not be altered by genuine give-and-take at each performance. That is, in her
many confrontations with Stanley, Blanche may sometimes seem to win and sometimes
to lose, but these small gains and losses will not lessen the overwhelming tragedy of her
being led off to the asylum in the last scene. In fact, there is an old theatrical saying:
'The only way to keep a scene the same is to let it change.' Acting that is rooted in real
give-and-take insures that the performance retains the original vitality but rarely dis-
torts the play's overall meaning or impact.l
Directors vary in their reaction to this aspect of acting. Some embrace it because
of the exciting, unpredictable performances it produces, while others prefer the same
denotative meaning at every performance, accepting only minor differences in emo-
tional valance from one rendition to the next. Film directors frequently ask actors to
engage in completely different dramatic intentions on different talces, so they can
choose between them during the editing process. (For a discussion of acting expert-
ise, see Noice and Noice, 1997, 2001, 2002, 2006. For comparisons between acting
and jazz, see Noice et aI., 2002.)
Thus, it could be said that these arts lie on a continuum, with classical music on
one end (most interpretations planned in advance, with only emotional thrust vary-
ing), acting in the middle (some interpretations planned in advance but some new
meanings actually being created at each performance) and jazz at the other end (new
melodies and harmonies created at every performance). Of course, this comparison
 Noice et al.: Memorization by a jazz musician 67

only looks at a single aspect of these art forms. Their essential complexity is far too
great to be captured along one dimension.
In this article, we propose that the same principles of expert memory specified by
Ericsson and Kintsch (1995) that we have previously suggested lie behind actors' remark-
able retention of complex text and classical musicians' flawless rendering of western art
music, also apply to jazz, at least during initial learning of a piece from the printed score.2
We asked a jazz pianist to memorize an unfamiliar piece from a lead sheet and observed
his practice as he did so. We expected to find that this jazz musician, Weeclassical musi-
cians, would use the formal and harmonic structure of the music to guide his practice
and memorization, and would prepare performance cues to guide him.
The pianist was asked to practice the music he was given until it was ready for per-
formance. Because the goal of the present study was to understand memorization, we
limited our observations to the initial preparation and did not observe finished, impro-
vised performances. The musical structure and performance cues that the pianist
practiced during his initial work on the piece would, however, provide the basis for
later improvised performances. Depending upon how much or how little true compos-
ition3 occurs during improvisation, decisions about interpretation and technique
(e.g. the positioning of the hand to start on the proper finger for executing a scale
over an anticipated chord progression) may have to be made in advance and perform-
ance cues prepared for their implementation. If they are to guide the musician's play-
ing during performance, the musical structure and performance cues must be
committed to memory and their retrieval practiced during the preparation of a new
piece for performance. The present study, thus, examined the use of structure and
performance cues by a jazz musician learning the note-far-note version of a new piece
(prior to improvising), adapting the methods used by Chaffin and colleagues for study-
ing the practice of musicians in the classical genre.

Method
A freelance jazz pianist videotaped his practice as he learned a new piece that he had
never seen before. In answer to a request for his background he supplied the following:
I grew up in a home with a father who is a jazz musician (vibes/piano), so I was sur-
rounded with jazz. He did not push me into playing, but in the summer between
third/fourth grade, after I showed interest by teaching myself to read treble clef and play
basic chords and blues progressions, he connected me with a great pianist in Chicago.
I mainly played classical pieces, but my teacher sketched (sometimes in less than a
minute or two) simple stride piano arrangements of standard tunes. So, I was introduced
into jazz and some underlying theory in addition to developing technique needed in both
classical and jazz piano. Later, I took up bass; in fact worked my way through college
mainly as a bassist because of the demand, but I never stopped playing jazz piano, which
accounts for most of my gigs these days. I've been playing professionally for well over
20 years.

The experimental instructions emphasized that the pianist should learn the selection
as if preparing for a solo performance. The pianist learned the piece in two videotaped
practice sessions, 10 days apart (one 15 minutes, and one 30 minutes), during which
he simultaneously verbalized his thoughts. The piece was Funk in Deep Freeze by
68 Psychology of Music 36(1)

Hank Mobley, an infrequently played selection of the bebop repertoire. The score was
notated in two eightbar sections with repeats in AABA form, including variations such
as second endings, common in lead sheet musical notation.
After learning the piece, the pianist provided reports about musical structure and
about decisions about technique, interpretation and performance by marking the fea-
tures of each dimension reported on a copy of the score. The pianist reported the loca-
tion of section, subsection and phrase boundaries, technical difficulties, and the basic
and expressive performance cues he relied upon to direct his playing of the piece. These
musical features and performance cues were coded with dummy variables representing
the start and end of each subsection, the start of each phrase, and location of each
basic and expressive performance cue. The pianist also rated the technical difficulty of
each bar on a la-point scale.
The pianist's comments during practice were transcribed and classified by topic
into nine categories (see Table 1). The classification was done independently by two
judges with 85 percent agreement and differences were reconciled by discussion.
Practice was transcribed by recording the locations in the piece where the pianist
started and stopped, and by counting the number of starts, stops and repetitions for
each bar. These starts, stops and repetitions served as dependent measures in mul-
tiple regression analyses in which the predictor variables were the musical features,
performance cues and ratings reported by the pianist. Preliminary analyses included
all predictors, but only those predictors that had significant effects were included in
the analyses of practice reported below.

Results and discussion

The pianist's practice (shown in Figure 1) was similar to that of classical musicians
in following a run-work-run pattern in which episodes of work (repetition of short
segments) alternated with longer runs that tied the shorter segments together
(Chaffin et aI., 2002; Miklaszewski, 1989).
Session 1 began with an uninterrupted run through Al and A2 (bars 1-16). The
pianist then worked on these sections until the end of the session, which concluded
with a final run that continued through E, previewing the work that remained to be
done. Session 2 began in similar fashion with a run through Al followed by work on
Al and A2 concluding with a run through Al and A2. Next, the pianist worked on E,
concluding with a run that tied AI, A2, E, and AI' together for the first time, pro-
viding the first (albeit interrupted) performance of his preferred form for the piece.
The second half of session 2 was devoted to additional practice performances, inter-
spersed with work on the A and E sections, and concluded with another (interrupted)
run through the whole AAEA form. At the end of session 2, the piece was almost
ready for performance. The pianist said: 'So, I'm feeling it's pretty much there. It's just
a couple of things ... [I need] about another five minutes.'

PIANIST'S COMMENTS
The participant's comments during practice indicated that he was concerned in
approximately equal measure with issues of performance structure, technique and
metacognition. Across the two sessions, each of these four main topics accounted for
TABLE 1 The percentage of comments on structure, technique, and performance in sessions 1 and 2
29
1113
33 1
11
12
320
5Session
19922 982 IOkay,
16Session
342026
25 I'm
just
What
Two
So,
If
Thehave 2just
I 21I'mAhave
Ito
were trying
sections,
usually
feeling
like to
toughest Example
to memorize
atoimprovise
really
like
deep
figure do
partthen
out work
isit's
freeze Session
analyze
theiton
the on this
pretty
goes
this
rhythmathe1first
I melodic
into
fingeringmuchchord line 14
B, line
thisthere
would
on this 20157 here
,13
10
progressions
,section
one line Percent comments
Harmonic
Evaluation
Rhythm
Memory

~
8'
Cl>
..•.

~
3o
~.
;:;
0'
;::s

&'
:>:l

~'

3i'::
~,
<>
!is'
;::s

0"\
I.D
70 Psychology of Music 36(1)

150

125

~ 100
0>

e
IOJl

'"
0>

~ 75
-.a
'"
co
•..
p.. 50

25
--
1

1 4 8 12 16 20 24 28 32 36 40 44
Bars

PIGURE 1 Practice graphs generated during sessions 1 and 2.

20-30 percent of the comments (Table 1). Memorization was the main performance
issue, especially in session 1. Memorization was also behind the focus on musical
structure, which was also more pronounced in session 1. The link between structure
and memorization is apparent in the following series of comments at the beginning
of session 1.
Okay ... Like a deep freeze ... so it's an F minor blues type tillng ... It's basically a ...
typical eight-bar two A sections and then it goes into a B section ... [then] back
to the beginning Typically with a ... minor blues [progression] you're always lead-
ing into the dominant, so that's easy to remember ... So, now I'm going to start mem-
orizing without even looking at the second part of the chart.

In these comments, the pianist appears to be identifying the harmonic and formal
structure, noting the link between structure and memorization, and deciding that his
understanding of the structure is sufficient to begin memorizing.
 Noiceet al.: Memorization by a jazz musician 71

The main difference between the two sessions was the increase in evaluative com-
ments in session 2 as the pianist continuously assessed the accuracy of his playing:
It's almost there. It's just not exact ... So what I'm trying to do is just iron out the lit-
I I
tle biases that have, how want to play it, versus how it's actually written ... That's
not what the composerwanted.
Memory and rhythm were the main issues. Rhythm presented technical challenges
in the piece.
As the pianist noted in session 2: 'This is really tricky ... because this is not a typ-
ical rhythm.'
Two sections of the piece in particular (bars 4-8 and 22-24) involved notes tied
across the bar line and characteristic bebop rhythms that posed technical problems.

PRACTICE
Practice was organized by the formal structure. We have already described the large-
scale organization of the practice shown in Figure 1: The pianist worked first on the
A sections and then on the B, and then on integrating them. The summary of the
regression analyses in Table 2 shows that practice was also organized by the formal
structure at a more micro-level. The pianist used the beginnings of subsections as
starting points in both sessions, starting on these bars 12.5 times more often than
other bars in session 1. The effect is evident in Figure 1, which shows that the pianist
preferred to start in some places more than others. For example, at the beginning of
session 1, after playing through the first 16 bars, the pianist began first at bar 1, then
at bar 3, then at bar 5. These are the beginnings of the first, second, and third sub-
sections respectively. The pianist reported that he divided A into two-bar subsections
and B into four-bar subsections based on harmonic progressions. For example, he
described bars 5-6 (the third subsection of A) as 'a typical cyclical pattern of cycle-
of-fourths with an inner chromatic pattern of tritones "buried" in the chords of the
cycle'. The pianist's use of harmonic structure to determine starting places through-
out his practice suggests that he continually had this structure in mind.
In session 1, the pianist also used the beginning phrases as starting points.
Phrasing was determined mainly by the location of rests, which provided a 'breath',
further subdividing the subsections. The absence of this effect in session 2 suggests
that by this time the pianist had integrated phrases into the larger unit of the sub-
section and no longer found it necessary to make so fine a subdivision of the piece.
The change reflects the increasing integration of smaller into longer practice seg-
ments already noted in the description of Figure 1.
The pianist started more at expressive performance cues in session 1 and started
and stopped more at basic performance cues in session 2 (see Table 2). The change can
be illustrated by comparing starts at bars 4 and 5, where the pianist reported respec-
tively a basic and an expressive performance cue, the latter annotated 'crescendo'. In
session 1, the pianist started eight times at the expressive cue in bar 5 and never at
the basic cue in bar 4. In session 2, in contrast, he used bar 4 as a starting point
27 times in addition to continuing to start at bar 5. The difference between the two
sessions suggests that the pianist began by establishing the expressive performance
 'J
N

'"\l
~
3-
<::>

~
.Q,
~
s:::
~.
n
W
9::.

TABLE 2 Regression coefficients for multiple regression analyses of effects of musical structure and performance cues on C
starts, stops, and repetitions in sessions 1 and 2
Predictor 126.070*
23.896
.205
-1.0101.054
.054
20.545
0.837
3.888
3.136
-0.3230.903
5.991
0.423
4.224
0.113
5.163*
18.143-
5.877*
5.436
.088
-1.141
2.953.264
2.956
3.498
2.300 1 1.604
0.523
Stops 12.546*
14.049~
9.669*
8.859**
3.779
Repetitions Starts
 Noice et al.: Memorization by a jazz musician 73

cues that determined the overall musical shape of the piece in session 1 and then added
the basic performance cues needed to remember exactly what to play in session 2.
Although practice in session 2 continued to be shaped by expressive cues, their effects
were obscured by practice of basic cues that were a more urgent concern. Therefore,
the effect of expressive cues was no longer significant.
There were no effects for technical difficulties whether coded with a dummy vari-
able or represented by the pianist's ratings. These predictors were, therefore, not
included in the analysis reported. Their absence is surprising, given that the tech-
nical difficulties all involved rhythm and that a substantial numbers of comments
were made about rhythmic difficulties in both sessions. Although the pianist was
clearly working on these difficulties, their effect was not reflected in the regression
analysis for two reasons. First, they were few in number. Second, the pianist reported
basic performance cues at the start of the rhythmically difficult passages and worked
on these, therefore contributing to the effect of basic performance cues in session 2.

Conclusion
The musician in the present study showed the same kind of attention to musical
structure and performance cues observed in earlier studies of experienced musicians
in the western classical tradition learning new works for performance (e.g. Chaffin
and Imreh, 2002; Ginsborg et al., 2006; Williamon and Valentine, 2002). Thus, the
principles of expert memory (Ericsson and Kintsch, 1995) appear to be applicable to
jazz as well as to musical performance in the classical idiom. Although the 45 minutes
of the practice in the present study was much less than in the studies of classical perform-
ers, the piece was correspondingly shorter and technically simpler. The amount of
practice and degree of preparation in the present study was roughly comparable to
that described by Chaffin (2007), who observed a concert pianist learning the 1607
notes of Clair de Lune for performance in four-and-three-quarter hours.
As in earlier studies of the practice of experienced musicians (e.g. Chaffin and Imreh,
2002; Miklaszewski, 1989, 1995), the jazz pianist used the hierarchical structure of the
piece to organize his practice, worldng on the piece in sections and using the beginnings
of subsections and phrases as starting points. Sections, subsections and phrases repre-
sented three levels in the hierarchical organization of the piece, with phrases nesting
within subsections nesting within sections. Interestingly, the pianist later pointed out
that he chose starting locations for practicing mainly because they provided convenient
points to get into rhythmically tricky phrases. This comment illustrates how musical
structure provides a context for the more deliberate problem-solving efforts that are the
conscious focus of a musician's efforts during practice. Although the pianist was not
explicitly thinking about structure, his choice of starting points shows that it, neverthe-
less, provided a framework for his work. In the same way, the structure of a journal
article provides a framework for a reader struggling to understand a particular argu-
ment. The reader does not normally think explicitly about the position of the argument
in the formal structure of the article, but the implicit knowledge that the argument is
part of the conclusion is important to understanding it.
In other studies of musicians, the formal structure has been shown to affect free recall
and hesitations in practice performances from memory in ways that provide additional
74 Psychology of Music 36(1)

support for this conclusion (Chaffin, 2007; Chaffin and Imreh, 2002; Williamon and
Valentine, 2002). We suggest that the effects of the formal structure on practice in the
current study indicate that the formal structure served the jazz pianist as a retrieval struc-
ture in the same way. Although it is beyond the scope of the present inquiry, we specu-
late that possession of a mental map of this sort may be essential for many aspects of
jazz improvisation (e.g. Pressing, 2000; Sudnow, 2001; but see also Berliner, 1994). That
is, the musical structure might provide a mental map that could be activated during per-
formance, allowing the pianist to keep track of where he is in the piece and, perhaps, to
decide where to go next.
The pianist reported that he used performance cues, paying attention to particular fea-
tures of the music during performance in order to remind himself of what came next.
He reported the location of the features that he used for this purpose, and the practice
data were consistent with his report. The data indicate that expressive cues served as
starting points for playing in session 1 and basic performance cues as starting and stop-
ping points in session 2. Starting at these points would have established them as retrieval
cues by linking the conceptual representation of the music in working memory with the
motions needed to perform it and the sounds those actions would produce.
The order in which expressive and basic cues were practiced is surprising at first
sight. One might expect basic issues of technique to be settled before nuances of
expression come to the fore. It is, however, a characteristic of experts' problem solv-
ing that they begin with 'the big picture' while novices jump into the details without
a clear idea of where they are going (Glaser and Chi, 1988). The expert approach to
learning a new piece of music is strongly recommended by the noted piano pedagogue
and pianist, Heinrich Neuhaus (1973) who urges the benefits of beginning work on
a new piece with a 'musical image' for how it should sound in mind. Expressive per-
formance cues represent the main musical turning points of the piece and their
effects in session 1 suggest that the pianist in this study was following Neuhaus's dic-
tum. Similar effects have been observed in other longitudinal case studies of classical
performers learning new pieces (e.g. Chaffin et al., 2003).
The pianist explicitly acknowledged leaving technical details until later in his com-
ments in session 1:

Typically what I do is make up my own line for complicated syncopated lines ... Mainly
I am concerned first with 'Do I understand the harmonies?' ... So I'm purposely just sort
of ignoring the incorrect playing of the melodic line.

However, it was not until the end of session 2 that he explicitly referred to his musical
image for the piece, using color as a metaphor: 'I think of most of this as purple with
some lighter blue sections.' In a post experiment interview, he expanded on this by say-
ing that he thinks of purple as the kind of music one might hear in a horror film.
The effect of basic performance cues on practice in session 2 but not in session 1 is
another indication that the pianist became more concerned with details in the second
session. The shift in focus was shown by session 2 comments reflecting his concern to
eliminate inaccuracies introduced earlier:

I'm hearing another song, that's why I want to jump to ... Let me try that again ... That's
right. And so now I'm feeling a lot more comfortable with this line, and it is just little tiny
pieces I have to iron out now. Let me just take that slow to make sure I'm doing that right.
Yeah, I was playing that rhythm wrong ... That's not what the composer wanted.
 Noice et al.: Memorization by a jazz musician 75

Basic performance cues reminded the pianist of the critical notes and rhythms needed
to play the piece as written.
In session 2, the pianist was also beginning to play through the entire piece in prac-
tice performances and needed to remember what came next as he moved from one pas-
sage to the next. Basic retrieval cues elicited the necessary information from memory.
For example, he described one of the basic performance cues in the following way:
'Playing the F in first subsection, Bl of section B, cues the phrase for the last half of
the B-section, B2.' Attention to basic performance cues in session 2 thus reflected the
pianist's concerns with both accuracy and playing from memory, reminding the pianist
of the critical notes and rhythms needed to play the piece as written.
It might be thought that memorization was simplified because many jazz standards
contain common progressions. According to the pianist, this was not the case with
Funk. He said:

It does not fit, in a full sense, any common (e.g. blues or I Got Rhythm) progression. The
first four bars (the first subsection of the A section is repeated which also makes it bars
9-12) follow generally the minor blues progression. However, the second subsection of
A diverges from this minor blues pattern and goes through a different common pattern
(cycle of fourths) which is the land-back at a turnaround used in a minor blues. However,
note that the A section is eight bars and minor blues (and blues in general) is 12 bars.
Moreover, the rhythmic pattern of when the next chord is played (and the length of time
one stays on that chord) is not common. The 'B' section has two subsections each, which
is a typical II-V-I pattern, but each is in a different temporary key from the original
F minor (first, B flat major; second, D flat major). I had to memorize what these tempo-
rary keys were. They are related to F minor/ A flat major in an indirect way, but I still
had to 'hear it in my mind' and memorize what these keys were before I could apply the
'common' chord progression pattern of II-V-I. Again, rhythm had to be memorized also.

In sum, the pianist noted that memorization was not materially aided by commonal-
ities with standard tunes.
It is possible that if practice had continued, the pianist's attention would have
returned to the musical image and to expressive performance cues as happened with
classical musicians whose practice was followed to the point that they were ready to
perform in public (Chaffin et aI., 2002). In the present study, however, practice did
not extend past initial memorization because available studio time was such that the
pianist did not quite reach the level where he was ready for the first performance.
It appears that the psychological processes responsible for jazz and classical mem-
orization may be very similar. That is, from a cognitive perspective, the differences
between performance practice in the two genres may be relatively superficial. In both,
virtuosity, creativity, and musical expressivity are highly valued. The main difference
is that in jazz, the ability to improvise in fresh and interesting ways is of prime import-
ance, while in the classical tradition, a much narrower range of variation across per-
formances is allowed. Even in the classical tradition, however, differences between
performances are the norm and most performers appear to share the view of Emil
Gilels, the noted Russian pianist, that such variation reflects well on the artist:

When I am in top form ... the ideas are always different. Sometimes I play with greater
changes in dynamics, sometimes with less ... I must say it is different each time I play,
and it is a process which ... includes mastery of the work, knowing the details, being
comfortable with it, and then adding the fantasy. (Mach, 1991; 123)
76 Psychology of Music 36(1)

However, these changes from performance to performance are very minor compared
to those made during jazz playing. The difference between idioms is captured by this
quote from jazz legend Keith Jarrett (who has also made many acclaimed classical
recordings). He said of his solo jazz performances:
When I go on stage, if I have an idea in my head, it's going to be in my way. Those notes
and feelings come to the player, come to the improviser, if he lets them. But if there's an
idea in the way, those notes and those feelings will be restricted to whatever that idea
started to be.

In the same interview, he described the process of improvisation as: 'It's like you're under
shock all the time - an electrical current is flowing through you' (Lehrer, 2005).
Despite the performance differences, it appears that the mental processes respon-
sible for memorization are very similar in the jazz and classical traditions. This is not
surprising given that similar principles account for expert memory in domains that
are much more different than jazz and classical music performance, such as chess,
acting, dance and waitering (Ericsson and Kintsch, 1995; Noice and Noice, 1997,
1999,2006). With respect to memory for note-for-note rendition, differences between
preparation for performance in the jazz and classical tradition appear relatively minor.
Both require the performer to have a conceptual representation of the piece with a
highly practiced retrieval scheme based on the musical structure. Both require use of
basic performance cues to ensure that critical notes are played as planned. Therefore,
the starting point for the jazz performer studied here and the classical performers stud-
ied previously appears to be very similar.4
The pianist in this study practiced until he was nearly ready to perform. Because
our goal was to understand memorization, we did not try to observe finished, impro-
vised performances. The practice that we observed did, however, provide the basis that
could be used for later improvised performances. The structural and performance cues
whose preparation we have described proved to be an effective memory retrieval
organization and provided the framework needed for improvisation. The spontaneity
of jazz performance requires suitable preparation. Depending upon how much or how
little true composition occurs during improvisation, decisions about technique
(e.g. playing over-the-bar phrases), and interpretation have to be made in advance
and performance cues prepared for their implementation. If they are to guide the
musician's playing during performance, the musical structure and performance cues
must be committed to memory and their retrieval practiced during the preparation of
a new piece.

NOTES

1. This discussion concerns only theatre in its most formal sense: a production in which the
actors perform a playwright's script from memory. In improvised theatre, the actors are
the playwrights, creating dialogue on the spot, constrained only by audience suggestions
or certain stated 'rules of the game'.
2. New music may also be learned from listening to recordings or live performances, but this
experiment looked at acquisition from a printed lead-sheet.
3. This musician uses the phrase, true composition, to refer to occasions where he surprises
himself by playing something he does not recognize as anything he had ever played or
heard before - something of which he says, 'I have no idea where that came from.'
 Noice et al.: Memorization by a jazz musician 77

4. This separation into jazz and classical musicians is for clarity and is intended to indicate
the primary field of an artist. Of course, many players are adept at both types and go back
and forth between fields.

ACKNOWLEDGMENTS

We gratefully acknowledge the assistance of James Pelletier in scoring and entering the data.
Portions of this article were reported at the Annual Meeting of the Psychonomic Society,
Minneapolis, November, 2005.

REFERENCES

Berliner, p.p. (1994) Thinking in Jazz: The Infinite Art of Improvisation. Chicago: University of
Chicago Press.
Chaffin, R. (2007) 'Learning Clair de Lune: Retrieval Practice and Expert Memorization',
Music Perception 24: 377-93.
Chaffin, R. and Imreh, G. (1997) "'Pulling teeth and torture": Musical Memory and Problem
Solving', Thinking & Reasoning: Special Issue on Expert Thinking 3: 315-36.
Chamn, R. and Imreh, G. (2001) ~ Comparison of Practice and Self-report as Sources of
Information about the Goals of Expert Practice', Psychology of Music 29: 39-69.
Chaffin. R. and Imreh, G. (2002) 'Practicing Perfection: Piano Performance as Expert
Memory', Psychological Science 13: 342-9.
Chaffin, R., Imreh. G., and Crawford, M. (2002) Practicing Perfection: Memory and Piano
Performance. Mahwah. NJ: Erlbaum Associates.
Chaffin, R.. Lemieux. A. and Chen. C. (2007) "'It is Different Each Time I Play": Variability
in Highly Prepared Musical Performance'. Music Perception 24: 455-72.
Chaffin. R.. Imreh, G., Lemieux. A. and Chen, C. (2003) "'Seeing the Big Picture": Piano
Practice as Expert Problem Solving', Music Perception 20: 461-85.
Chase, w.G. and Ericsson, K.A. (1982) 'Skilled and Working Memory', in G.H. Bower (ed.)
The Psychology of Learning and Motivation. pp. I-58. New York: Academic Press.
Ericsson, K.A. and Kintsch, W. (1995) 'Long-term Worldng Memory', Psychological Review
102: 211--45.
Gabrielsson. A. (1999) 'The Performance of Music'. in D. Deutsch (ed.) The Psychology of
Music (2nd edition). pp. 579-602. San Diego. CA: Academic Press.
Ginsborg, J.. Chaffin, R. and Nicholson. G. (2006) 'Shared Performance Cues in Singing and
Conducting: A Content Analysis of Talk during Practice'. Psychology of Music 34(3): 167-94.
Glaser. R. and Chi, M. (1988) 'Overview', in M. Chi. R. Glaser, and M. Parr (eds) The Nature
of Expertise. pp. xv-xxviii. Hillsdale, NJ: Erlbaum.
Hallam. S. (1995a) 'Professional Musicians' Approaches to the Learning and Interpretation
of Music', Psychology of Music 23: 111-28.
Hallam. S. (1995b) 'Professional Musicians' Orientations to Practice: Implications for
Teaching', British Journal of Music Education 12(1): 3-19.
Hallam. S. (1997) 'The Development of Memorization Strategies in Musicians: Implications
for Education'. British Journal of Music Education 14(1): 87-97.
Hallam. S. (2001) 'The Development of Metacognition in Musicians: Implications for
Education'. British Journal of Music Education 18(1): 27-39.
Imreh. G. and Chamn. R. (1996/97) 'Understanding and Developing Musical Memory: The
View of a Concert Pianist'. American Music Teacher 46(3): 2()-4.
78 Psychology of Music 36(1)

Kenny, BJ and Gellrich, M. (2002) 'Improvisation', in R. Parncutt and G.E. McPherson (eds) The
Science and Psychology of Music Performance, pp. 117-34. New York: Oxford University Press.
Lehmann, A.C. (2005) 'Komposition und Improvisation: Generative musikalische
Performanz', in T.H. Stoffer and R. Oerter (eds) Allgemeine Musikpsychologie, pp. 913-54.
Gottingen: Hogrefe.
Lehmann, A.C. and Ericsson, K.A. (1998) 'Preparation of a Public Piano Performance: The
Relation between Practice and Performance', Musicae Scientiae 2(1): 67-94.
Lehmann, A.C. and Gruber. H. (2006) 'Music', in. K.A. Ericsson, N. Charuess, P. Feltovich,
and R. Hoffman (eds) Handbook on Expertise and Expert Performance, pp. 457-70.
Cambridge: Cambridge University Press.
Lehrer, J. (2005) Interview with Keith Jarrett, The Newshour with Jim Lehrer Broadcasts,
19 July. New York and Washington, DC: Public Broadcasting Service.
Mach, E. (1991) Great Contemporary Pianists Speak for Themselves. Mineola, NY: Dover Books.
Mildaszewski, K. (1989) L\. Case Study of a Pianist Preparing a Musical Performance',
Psychology of Music 17: 95-109.
Miklaszewski, K. (1995) 'Individual Differences in Preparing a Musical Composition for
Public Performance', in M. Manturzewska, K. Mildaszewski, and A. Bialkowski (eds)
Psychology of Music Today: Proceedings of the International Seminar of Researchers and
Lecturers in the Psychology of Music, pp. 138-47. Radziejowice, Poland.
Neuhaus, H. (1973) The Art of Piano Playing. New York: Praeger Publishers.
Noice, 1~ and Noice, H. (1997) Expertise of Professional Actors: A Cognitive View. Hillsdale, NJ:
Lawrence Erlbaum Associates.
Noice, H. and Noice T (1999) 'Long-term Retention of Theatrical Roles', Memory 7(3): 357-82.
Noice, H. and Noice, T. (2001) 'Learning Dialogue With and Without Movement', Memory
and Cognition 29(6): 820-8.
Noice, 1~ and Noice, H. (2002) L\. Review of Recent Research on the Expertise of Professional
Actors', High Ability Studies 13(1): 7-19.
Noice, H. and Noice, 1~ (2006) L\.rtistic Performance: Acting, Ballet, and Contemporary
Dance', in A. Ericsson, N. Charness, P. Feltovich and R. Hoffman (eds) Handbook on
Expertise and Expert Performance pp. 489-504. New York: NY: Cambridge University Press.
Noice, H., Noice, T. and Jeffrey, J. (2002) 'Mental Representation of a Jazz Musician', poster
presented at the 43rd Annual Meeting of the Psychonomic Society, November, Kansas
City, MO, USA.
Pressing, J. (2000) 'Improvisation: Methods and Models', in J.A. Sloboda (ed.) Generative
Processes in Music: The Psychology of Performance, Improvisation, and Composition,
pp. 129-78. New York: Oxford University Press.
Shaffer, L.H. (1984) 'Timing in Solo and Duet Piano Performances', Quarterly Journal of
Experimental Psychology 36A: 577-95.
Shaffer, L.H., Clarke, E.P. and Todd, N.P. (1985) 'Meter and Rhythm in Piano Playing',
Cognition 20: 61-77.
Sloboda, J.A. (1994) 'Music Performance: Expression and the Development of Excellence, in
R. Aiello and J.A. Sloboda (eds) Musical Perceptions, pp. 152-169. London: Oxford
University Press.
Sudnow, D. (2001) Ways of the Hand: A Rewritten Account. Cambridge MA: MIT Press.
Ward, G. (2000) A History of America's Music. New York: Alfred Knopf.
Williamon, A. and Valentine, E. (2002) 'The Role of Retrieval Structures in Memorizing
Music', Cognitive Psychology 44: 1-32.
Williams, T. (1947) A Streetcar Named Desire. New York: New Directions.
 Noice et al.: Memorization by a jazz musician 79

HELGA NOICE is a cognitive psychologist. She and her husband, Tony, have published dozens
of articles in such journals as Memory, Cognitive Science, and Memory & Cognition. They have
also conducted two NIA-funded arts interventions to enhance healthy cognitive aging. Helga
is a Professor of Psychology at Elmhurst College.
Address: Department of Psychology, Elmhurst College, 190 Prospect Ave., Elmhurst, IL 60126,
USA. (email: helgan@elmhurst.edu)

TONY NOICE has been a professional actor/director for over 35 years. He teaches acting at Elmhurst
College, and, along with his wife, Helga, researches the cognitive processes involved in performance.
Together, the Noices have published dozens of articles and books in the professional literature.
Address: Department of Communication Arts and Sciences, Elmhurst College, 190 Prospect
Ave., Elmhurst, 1L 60126, USA. (email: noicea@elmhurst.edu)

ROGER CHAFFIN is Professor of Psychology at the University of Connecticut. His research on

musical memory has appeared in journals such as Psychological Science, Music Perception, and
Music Psychology and in the co-authored book, Practicing Perfection: Memory and Piano
Performance (Erlbaum, 2002). His work on memory and language appears in numerous journal
articles and two books.
Address: Department of Psychology, University of Connecticut, 1800 Asylum Avenue, West
Hartford, CT 06117-2697, USA. (email: Roger.Chaffin@uconn.edu)

JOHN JEFFREY has been a professional musician (piano and bass) for over 25 years, playing
solo or with jazz groups literally hundreds of times a year in the Chicago area. In addition to
his music career, John is a computer scientist, having earned his PhD from the University of
illinois at Chicago. He teaches programming at Elmhurst College.
Address: Department of Computer Science and Information Systems, Elmhurst College, 190
Prospect Ave., Elmhurst, IL 60126, USA. (email: johnj@elmhurst.edu)