![[PDF]](/pb-assets/Icons/adobe-pdf-icon-1498649896243.png){kind=icon}
Abstract
This study investigated the routine procedures employed by nine undergraduate piano students at a Brazilian university while learning and performing memorized pieces and the procedures employed using Chaffin’s performance cue (PC) protocols. The data were collected in two phases. In Phase I, each participant selected one piece that he or she had previously studied and memorized in the preceding academic semester. For Phase II, the students were introduced to Chaffin’s PC protocols, and they were allowed 10 weeks to memorize another piece of their own choosing. Thereafter, a second semi-structured interview with specific criteria was carried out, and the performance of the memorized piano piece was recorded. In Phase I, there were frequent references to structural and expressive cues. In Phase II, the employed PCs were shown to be related to the nature of the style of each piece, which in turn may indicate that explicit memory (content-addressable cues) seems to be associated with the deliberate expression of a given piece’s stylistic structure. Furthermore, tempo also seems to modulate the frequency of the PCs necessary to guarantee a successful memorized performance, for example, a faster tempo results in fewer PCs being employed.
Introduction
Playing an instrument and/or singing requires the reconstruction of sounding objects for motor production, whether played by memory or not. In the specific case of a memorized performance after practicing a musical work in the Western classical tradition, a retrieval system made up of explicit and implicit patterns that were previously stored is required, and such a phenomenon is dependent on expertise level. According to Williamon and Egner (2004), the effective use of highly ordered retrieval schemas for memorizing music develops as a function of expertise.
Recent literature has presented studies on memorized performances with novices and with professional musicians. It is safe to say that the former tend to use the so-called rote memory; inexperienced music students usually repeat the musical material until certain interconnected chunks are created (forward chaining). Each chunk functions as a cue for the next, so that the playing of one chunk functions as a trigger for the next one (Lehmann, Sloboda, & Woody, 2007). If all goes well, the instrumentalist successfully reaches the end of the performance. However, the connection between chunks is often broken due to various factors, such as anxiety, extraneous thoughts, poor learning habits, or irregular practice routines; these factors are the most common causes of failure. We all know how frustrating it is to grope for some shred of data that will allow a performance to resume. As frequently discussed in the literature (Aiello & Williamon, 2002; Hallam, 1997), novices and relatively inexperienced players are more prone to suffer from this problem than are seasoned performers.
According to Aiello and Williamon (2002), several pianists and piano pedagogues have tried to systematize the ways in which performers learn music when preparing for a memorized performance. Three general approaches can be identified, namely: aural, visual, and kinesthetic approaches. The aural approach enables individuals to imagine sounds and to evaluate the performance’s progress. Visual memory consists of images of the written score and other aspects of the environment; for instance, a pianist may remember the positioning of his or her hands and fingers or the look of the chords as they are struck. Finally, kinesthetic memory (i.e., finger, muscular, or tactile memory) allows performers to execute complex motor sequences automatically.
Several significant studies have provided evidence to indicate that experienced performers approach memorization early in the learning process and devise memorizing strategies that include individually designed procedures, such as writing down parts of the piece, analyzing the music away from the instrument, starting in different places in the piece or singing one voice while playing another—the list affords many possibilities. In line with Lehman and collaborators, these procedures lead to the creation of a web of relationships that can be understood as meaningful units (chunks) organized in a tree-like structure (Lehman et al., 2007). Experienced performers seek to establish chained representations of the piece that are deliberated upon to become associated with specific physical actions and commands (Aiello & Williamon, 2002). This process allows information to be encoded, and retrieval takes place through meaningful and well-learned structural components and knowledge-based associations that connect patterns and schemas (Chaffin & Imreh, 2002; Ericsson & Kintsch, 1995; Krampe & Ericsson, 1996), that is, through music content-addressable memory (Chaffin, Logan, & Begosh, 2009).
It is likely that strategies for implicit memorization are subject to training, as are practice skills in general. The use of explicit and implicit memories seems to be an indispensable requirement for the performance of music. Incidental (implicit) memory has been shown to be superior for retrieving relevant information for representative tasks in domains such as sports, ballet, music and medicine due to deliberate practice (Ericsson, 2008). Deliberate memorization transforms the motor and auditory chains into discrete chunks of material that can be recalled individually, both through associate chains and content-addressable memories, which are smoothly integrated by musicians (Chaffin, Logan, et al., 2009). Retrieval in this modality requires some previous steps from which the information can be accessed, not from a content-less address but instead from some feature of the content itself, which is used to retrieve the remainder of the information. For instance, a single fact can be used as a marker to recall other related details, which results in the reconstruction of an entire pattern of information. This type of memory has the advantage of allowing greater recall flexibility because it is more robust. This distributed memory is able to work around errors by reconstructing information that may have been hidden or scattered throughout the system.
Since 1997, Chaffin and collaborators have investigated the memorization strategies employed by professional musicians when preparing for performances. Their findings have generated a conceptual scaffold that embraces several interconnected dimensions of understanding and structuring musical events. Along with performers, researchers and performer-researchers, Chaffin proposes a systematization of categories, designated as performance cues (PCs), that may be employed by musicians to monitor the unfolding performance and to adjust its automatic motor sequences to the needs of the moment (see, for instance, Chaffin, 2011; Chaffin & Imreh, 1997, 2002; Chaffin, Imreh, Lemieux, & Chen, 2003; Chaffin, Lemieux, & Chen, 2007; Chaffin, Lisboa, Logan, & Begosh, 2010).
Performance cues are landmarks that professional musicians employ during memorized public performances (Chaffin, Logan, et al., 2009). Performance cues suggest a deep and systematized knowledge that organizes a score into different categories: (i) basic cues are related to specific technical aspects (e.g., fingerings, physical posture, breathing and the resolution of basic mechanical issues); (ii) structural cues, similar to punctuation in written texts, are related to aspects of the formal structure, such as cadences, section endings, changes in design and musical boundaries; (iii) interpretative cues are related to musical gestures required to assure attention to dynamics, pacing and contour; and (iv) expressive cues are related to extramusical feelings, such as affect, atmosphere and images perceived and conceived by the interpreter. The systematization of these landmarks of mental representation has been proposed as a protocol for delimiting musical aspects and parameters to afford close and systematic references within the work. The performer can plan his/her practice and performances through a series of steps aimed at specific goals that are set during the learning process.
Seasoned musicians tend to develop systems of landmarks and maps that may, to some extent, ensure a resilient web of retrieval. The processes of deliberating about and ascribing markers and landmarks to memorized musical material were examined by Chaffin and his collaborators. The researcher and his team proposed a system of performance cues known as the PC protocol. Recent examples of PC use in memorized practices or performances have been reported in the literature, particularly as applied by professional musicians (Chaffin et al., 2010; Lisboa, Chaffin, & Demos, 2013; Lisboa, Chaffin, & Logan, 2009; Lisboa, Chaffin, Logan, & Begosh, 2007; Noice, Jeffrey, Noice, & Chaffin, 2008). The use of Schenkerian analysis in relation to PCs has also been employed to solve the problems of memorizing while developing a clear idea of the “big picture” (Chaffin, Gerling, Demos, & Melms, 2013).
In the literature, a few studies have researched students’ PC usage. Chaffin, Demos, and Crawford (2009) investigated a BA-level trumpet performance major and a 14-year-old piano student. Lisboa et al. (2013) described how an 18-year-old piano student learned to memorize by recording her thoughts, marking them on copies of the score.
Memorization had not been researched during our previous investigations related primarily to how students prepared new musical work without guidance from instructors (Gerling & Santos, 2011). Our previous conclusions showed that students were capable of expressing some parameters, namely, melodic contour and global coherence, at the expense of timing, articulation, character and tempo; in other words, interpretative choices were incipient at best. Our current research question addresses the potential role of Chaffin’s protocol in motivating students to manipulate and command a broader scope of interpretative parameters during their performances. Studies have shown that PCs have been employed to improve memorized performance (Chaffin et al., 2013). To the best of our knowledge, few if any systematic studies have been conducted at the apprentice level in relation to PCs as tools or support for highlighting some specific interpretative aspects during the preparation of instrumental pieces. It is our contention that PCs, acting as a framework for structured memorization, can improve students’ awareness of potential interpretative choices and can generate integrative habits in performance practices. In the present research, which aims at evaluating the effect that the PC protocol has on students’ interpretative choices, we initially investigated the unprompted routine procedures employed by piano majors for their memorized performances to map the nature of their memorization procedures and thoughts. Such results may provide additional aspects that may have been neglected in the initial conception of PCs. Furthermore, the use of PCs by a large number of students from a wide range of academic levels may provide an entire set of complementary information regarding how undergraduate music students proceed during memorized performances when using PCs. It has been our aim to add to the results already reported in the literature, as prior studies mostly involve professional musicians.
Method
Participants
Nine undergraduate piano students from a Brazilian federal (public) university participated in this study (seven males, two females; mean age = 21.5 years, SD = 2.8, age range = 18–27 years). Each student had approximately 9.3 years of formal music learning (SD = 1.3 years, range = 8–10 years). The following labels were employed: U represents an undergraduate (college-level) student; the number following the letter represents the academic rank and corresponds with the academic semester; and the lower case letter that follows (a, b, c) represents the individual participant.
Materials
In Phase I, the students presented the following pieces: (i) Chopin’s Etude Op. 10, no. 2 (U2a); (ii) Mozart’s Sonata K 333, 1st movement (U2b); (iii) Villa-Lobos’ Valsa da dor (U4a); (iv) Beethoven’s Sonata Op. 27, 1st movement (U4b); (v) Respighi’s Prelude I (Tre Preludi sopra melodie gregoriane) (U4c); (vi) Villa-Lobos’ The Porcelain Doll (Branquinha) (U4d); (vii) Santoro’s Piano Sonata no. 3, 2nd movement (U6); (viii) Chopin’s Impromptu Op. 36, no. 2 (U7); and (ix) Schoenberg’s Op. 11 (U8).
In Phase II, the students played the following pieces: (i) Chopin’s Scherzo Op. 31, no. 2 (U2a); (ii) the Exposition of Mozart’s Sonata KV 333, 3rd movement (U2b); (iii) the Exposition of Haydn’s Sonata (Hob. XVI: 37), 1st movement (U4a); (iv) Debussy’s Minstrels (U4b); (v) the Exposition of Beethoven’s Sonata Op. 110, 1st movement (U4c); (vi) the Exposition of Op. 81 Beethoven’s Sonata, 1st movement (U4d); (vii) Chopin’s Ballade Op. 47, no. 3 (U6); (viii) the Exposition of Mozart’s Sonata KV 332, 1st movement (U7); and (ix) the Exposition of Beethoven’s Sonata Op. 2, no. 3 1st movement (U8).
The seemingly wide variety of pieces featured in both phases can be justified by the fact that we chose not to interfere with the students’ previously assigned repertoire during the academic semester. In addition, this condition allowed us to further observe memorization under different music styles.
Procedure
Figure 1 depicts the sampling procedure employed during the data collection for both phases, occurring over 15 weeks within the academic semester. In the first week, the researchers invited students who belonged to two piano laboratories (collective classes in which students have the opportunity to perform, discuss and be evaluated by colleagues and piano teachers) to participate under the supervision of three piano teachers. At this time, the aims of the study and the data collection schedules were explained to the potential participants. From these laboratories, 12 out of 20 students opted to participate in the research, but only 9 continued until the end of the sampling.
Procedures in Phase I
In Phase I, each participant selected one piece that he or she had previously studied and memorized in the prior semester. The students were asked to review the pieces over two weeks. During this phase, students did not receive instructions nor any type of information regarding memorization. Students were informed of the importance of relying on their own resources, that is, the methods and procedures that they had typically individually employed during their practice sessions.
Students’ statements were collected during semi-structured interviews aimed at identifying their seemingly unstructured thoughts and memorization resources and devices. Sampling data also included recordings of performances and the semi-structured interviews. The following aspects were registered: (i) personal information (age, academic semester, piano teacher, length of formal piano learning, and selected piece/section); (ii) memorization routine procedures (the most frequently employed procedures; the strategies, if any, specifically used for this research; the places that they felt most sure during this particular performance; and the contexts of memorized public performance); (iii) memorization failures and near failures during the performance; (iv) the ability to start at any point in the piece; (v) the types of thoughts that occurred during memorized performance; and (vi) the style and tempo (perceived differences in memorizing Bach, Mozart or contemporary pieces; slow or fast tempo).
Procedures in Phase II
For Phase II, in week 4, we requested that the participants select one of the new pieces that they were supposed to prepare as part of their assigned repertoire for the academic semester. At this point in time, all participants received a thorough explanation of Chaffin’s protocol. They received a printed 10-page Portuguese translation detailing the concepts and procedures related to the protocol. R. Chaffin provided the materials to be translated and encouraged their use at our school. The students received permission and encouragement to make as many copies of the scores as they felt necessary. The students were also instructed to number the measures and to devise a color scheme for each of the four PCs categories.
After 10 weeks, the participants were asked to record their memorized performances. At this time, they were also asked to present the scores that they had used to practice and their annotations of PCs. The observations were extracted from the PCs marked after each practice session and each recording session to check the memorization level and to register the adopted tempo. Following their memorized performance, students were asked to mark the PCs that they used on a clean score. Thereafter, a semi-structured interview was conducted that investigated the following aspects of Phase II: (i) the perception of the role the PC protocol played in the memorized performance; (ii) the types of PCs employed (asked immediately after the performance); (iii) the use of cues other than those suggested in Chaffin’s PCs; (iv) the places where PCs were applied, including possible points of arrival and points of departure.
The interviews were transcribed and shown to the students to allow them to check, confirm or omit parts of their statements. There were no requests for omissions. All participating students signed an informed consent document.
Data analysis
The data were then transcribed and analyzed using interpretative phenomenological analysis (IPA) (Smith & Osborn, 2003), which aimed to explore in detail how participants were making sense of their personal experiences, perceptions or accounts of an object or event (e.g., in the present study, the deliberate memorization of a musical piece). As stated by Smith and Osborn (2003, p. 63), “the participants are trying to make sense of their world; the researcher is trying to make sense of the participants trying to make sense of their world.” From this perspective, some categories were extracted by taking into account issues that seemed to be relevant during the interviews. Table 1 includes some excerpts from the interviews, as well as their interpretation in terms of employed procedures (categories).
|
Table 1. Interview excerpts and corresponding interpretative categories of routine memorization procedures.
The analysis of all the participants’ statements during Phase I (N = 9) revealed seven main categories:
mental (visualization of the score);
topography (focus on the hands on the keyboard);
kinesthetic hints (awareness of the direction of the hand movement);
auditory (sound direction);
structural (harmonic structure and sections);
technical (leaps and scales); and
expressive (metaphor/feeling).
For the second phase, the categories were extracted from the cues notated during the students’ practice and the cues notated immediately after the recorded performance. The annotations were classified in terms of the PCs categories (basic, structural, interpretative and expressive), tabled in terms of incidence numbers and then normalized by the total number of measures of the played section(s). Figure 2 illustrates a typical example of a student’s markings on the score. The students were instructed to use different colors to mark the different PC categories. In this example, black, lined, grey and white arrows represent basic, structural, interpretative and expressive cues, respectively. We obtained similarly marked scores from all participants.
Statistical analysis
The Statistical Package for the Social Sciences (SPSS for Windows, version 19, IBM®) was used to analyze the data relationships related to the frequency (incidence) of the detected/interpreted routine procedures (extracted from the interviews), the employed (marked scores) PCs (during practice sessions and after performances) and performance tempi (from the recordings). All statistical tests were performed at the P < 0.05 level of significance. A cluster analysis was applied to data standardized through a z-score transformation to avoid misclassification due to possible differences in data dimensionality. Standardization eliminates the influence of different units of measurement and renders the data dimensionless. The distances between samples were calculated using square Euclidean distances. The dendrogram similarity scales generated by the SPSS program range from zero (greater similarity) to 25 (lower similarity).
Results
Routine procedures for memorization
During Phase I, the mean performance of the selected piece lasted 3.16 ± 1.98 min (range 1.03–6.45 min; median 2.18 min). During the recording of performances, three out of nine students failed during the memorized performance. The interview lasted 10.16 ± 4.28 min (range 5.08–19.22 min; median 9.15 min).
The following aspects were extracted from the analysis: (i) memorization strategies; (ii) specific places that came to mind during the performance; and (iii) the specificities of this particular performance with respect to style and tempo features. Figure 3 depicts the thematic categories extracted from the interviews in the first phase in terms of practicing memorization and performing by memory. In terms of memorized performance, the first category consisted of the global focus of attention states for aspects usually discussed or mentioned during the interviews, without pointing to specific locations (measures, phrases, cadences or sections) in the score. The second category embraced specific aspects of focal attention related to the information provided and associated with specific locations in the score.
According to Figure 3, most of the aspects related to the practice of playing from memory are relatively general. Nevertheless, concerning the performances by memory, the categories that emerged are related to attention on both global and specific aspects.
The analysis of the responses given by the students shows a mix of categories employed during the memorization process. Figure 4 shows the distribution of these categories in terms of frequency.
According to Figure 4, structural cues (26%) were most frequently noted as a memorization procedure. It is worth mentioning that, for this sample, the correlation analysis of the frequency of the employed cues showed inverse Spearman correlations (–0.633) between auditory and kinesthetic cues at a p < 0.05 level, that is, students who employed kinesthetic cues did not seem to use auditory ones, and vice versa.
The Phase II results show that the participants used the four PCs roughly equally. No detectable memory lapses seemed to have occurred during this phase. The distribution of the four PCs—basic, structural, interpretative and expressive—are shown in Figure 5.
During the interviews, in terms of interpretative PCs, students mentioned features such as legato, staccato and dynamics. During Phase I, the students did not verbalize such parameters as a focus for memorization, although they may have used interpretative tools during the first phase. Considering that the students employed interpretative features during Phase II (52.5%), it seems that the students primarily focused on interpretative elements as part of their memorization routines. Figure 6 shows the normalized incidences of the four categories of PCs employed by the participants. The total number of incidences was corrected by the total number of bars in each performed section.
As shown in Figure 6, the second-semester students, that is, the group with the least academic experience, marked a smaller number of PCs on their scores. Equally, the most advanced student (U8) employed just structural and expressive PCs. The others (U4 to U7) employed a broad number and types of PCs. It is worth mentioning that three students (U4a, U4c and U6) seemed to focus on interpretative PCs. These results suggest that such students focused on parameters such as articulations, phrasing, contour and dynamics; in other words, they focused their attention on interpretive cues rather than on basic cues.
Figure 7 shows the relative percentages of the PCs categories employed by the participants. The relative percentage was calculated by considering the number of incidences of a given PC and the total number of employed PCs in a given case.
According to Figure 7, in terms of the categories of PCs employed, the preference for basic cues was relatively stable in this sample. Structural PCs were most used by first-year students (U2a and U2b). Structural PCs were, however, less prevalent as the chosen cues for the upper classmen in this sample. Student U8 reported that his/her basic and interpretative PCs had already been assimilated. Therefore, he/she felt that structural and expressive cues adequately represented his/her thinking at that particular time in his practice and memorization.
Figure 6 partially illustrates distribution of the categories of PCs employed by U7 for the exposition of the first movement of Mozart’s KV 332.
As shown in Figure 8, U7 marked several interpretative PCs at the beginning of the sonata. She seemed to be focusing on articulation, that is, the type of touch she hoped to achieve and eventually learn; thus, she used PCs as a learning tool. Although she did not write down any expressive cues, during the interview the participant emphasized her views on the relationship between Mozart’s instrumental music and opera, the singing of opera arias and the importance of transposing the singing character to her playing on the keyboard. According to her view, the right hand corresponds to the singing voice, and the left hand is the supporting orchestra. These results highlight a contrast between the PCs explicitly marked on the score and the verbalized intentions and thoughts during performance. Similar results have been observed in the literature (Lisboa, Chaffin, & Logan, 2011). Moreover, she stated that the basic category helped her with her own resolutions of the technical difficulties found in Mozart’s score. She shared routine phrasing procedures:
[When I am playing,] I am always thinking that a phrase has come to an end, and then a new one starts, which, in turn, is followed by another and another until the big final point.
It is then fair to assume that she used structural and expressive cues as meaningful memorization devices.
As shown in Figure 9, U4d used interpretative cues in close relation with the dynamic changes in the first movement of Beethoven’s Sonata Op. 81a. Dynamics seemed to be a very important parameter for this student. Basic cues, in turn, seemed to relate to his desire to be accurate at all times. During his interviews, he made sure that he mentioned the sound of the horns in the initial measures, and he continuously emphasized the importance of the woodwinds in creating the right atmosphere for the Introduction (ms. 1–16). He did not mention the pacing change following the Introduction, but m. 21, with the galloping of horses, constituted an important cue for him. U4d had clearly established a metaphoric, or expressive, cue.
The Phase II data allowed us to extract relationships with tempo and music style. Figure 10 depicts the relationship between tempi taken during performances and the total number of performance cues marked after the performances; a faster tempo meant that the students in this sample employed fewer PCs. In addition, the analysis revealed an inverse correlation between tempo and interpretative PCs (rSp = −0.751, for p ⩽ 0.05).
A cluster analysis considered the measured tempi in the performances and the frequency of each of the four PCs marked by each student. The data were then normalized by the total number of measures of the part/piece performed. The resulting dendrogram was interpreted in terms of similarity among the students. Interestingly, style similarities between the pieces emerge as a possible strong factor in this group, as shown in Figure 11.
Most of the participants used PCs that were shown to be related to the style of the piece, which, in turn, may indicate that explicit memory (content-addressable cues) is associated with the deliberate expression of a given piece’s stylistic features. For instance, the similarities between PCs ascribed to pieces belonging to the so-called classical period (and that includes Beethoven’s Op. 2) have been brought to our attention. Likewise, student PCs ascribed to two of Beethoven’s later works (Op. 81a and Op. 110) and Chopin’s Op. 47 also exhibited a great deal of similarity. It is worth noting that the sheer length of Chopin’s Op. 31 probably set it apart in this classification. In addition, in spite of there not being a minimum number of cases for applying cluster analysis (Hair, Black, Babin, & Anderson, 2013), one cannot neglect that this study depends on a small sample. Therefore, the results should be interpreted with caution. The results are based on an exploratory statistical technique. We can only hypothesize that a larger sample would perhaps confirm the trends observed here.
Furthermore, tempo also seems to modulate the frequency of the PCs necessary for a successful memorized performance, that is, a faster tempo results in fewer PCs being employed. In Phase I, we found no discernible relationships between composition tempi and styles and the nature of memorization procedures.
Discussion
In Phase I, regarding structural strategies, students seemed to be mostly concerned with the beginning and end of each section. In addition, seven out of nine students mentioned the importance of the harmonic context. None of the students described any type of behavior related to note-by-note memorization. On the contrary, the students seemed to attempt to establish relationships between events and to look for possible patterns. Lehmann and Ericsson (1997) found that pianists who attempted to memorize a short piece note by note were slower in committing the piece to memory than those who conceptualized the piece in terms of its harmonic and melodic structure. Nevertheless, none of the students mentioned the use of interpretative resources (such as dynamics, articulation, or phrase contours) for memorizing their repertoires in Phase I (i.e., their routine memorization procedures). Of course, we cannot neglect the possibility that the students may be using note-by-note or interpretative strategies without verbalizing them.
In Phase I, the students verbalized routine procedures that were either procedural and/or declarative ways of thinking and knowing. Two students (U7 and U4b, see Table 1) stated concerns about the keyboard’s topographic issues; both students spoke quite often about the ways in which they looked at the keyboard to develop specific pattern awareness. In addition, U7 highlighted the necessity of being aware of what occurred during her performance, for instance, the succession of events (what comes next). This type of thought demonstrates procedural (motor-based) knowledge of how to do something. The two other participants presented in Table 1 seem to have employed more explicit memorization processes. For instance, U6 seemed to be targeting some specific way of accessing explicit memory by selecting landmarks akin to PCs. He spoke of very specific spots in the score, and he confidently mentioned that he would eventually be able start from anywhere if the need arose. On the other hand, U4a not only claimed that she had to visualize the score, but she also referred to the character of the piece and the feelings that she attributed to particular passages. Although she based her retrieval on the feelings and expressions that she ascribed to the piece’s character, she was able to address her playing in an objective manner, and she could confidently begin from a variety of places in the score, thus showing her command of content-addressable memory. In the literature (Chaffin, Logan, et al., 2009), there are meaningful discussions of the relationship between associative chains and content-addressable memory regarding procedural (implicit) knowledge and declarative (explicit) knowledge. This research seems to confirm that “to memorize a piece of music for performance, the musician must smoothly integrate the two kinds of memory” (p. 352).
In Phase I, expressive concerns were second among the students’ preferences (c. 23%). Several students spoke about the existence of a given “atmosphere” (rather than the presence of a character). Indeed, it seems that expressive decisions along with symbolic attributions were part and parcel of their memorization routine procedures. It is worth noting that structure and expression were two of the four main categories proposed by Chaffin and collaborators (Chaffin, Logan, et al., 2009). At the same time, the students’ discourses tended to show a great deal of reliance on sensorial/perceptive aspects that are more commonsensical devices and gimmicks, such as the auditory and kinesthetic actions applied during their recall. As researchers, however, we cannot fail to notice that these students belonged to diverse studios and that each student’s teacher, perhaps aware of his or her student’s conduct, may have valued and reinforced particular techniques for learning and memorizing.
In fact, the teaching studio dynamic and the relationship between master and apprentice may define the adoption of the learning modes and behaviors that eventually translate into explicit and/or implicit cuing processes that may include the three other aspects emphasized by the students during Phase I, namely: auditory, topographic and kinesthetic memory (c. 14% each, as shown in Figure 2). According to Finney and Palmer (2003), auditory memory tells the musician what comes next, providing cues to elicit the music from memory, while also letting the musicians know that the piece is progressing accurately. In addition, in this master–apprentice relationship, kinesthetic cues are also stressed. During practice, students, after several repetitions, acquire motor programs that become part of the body’s motor memory. According to Ebert, Deller, Steffen, and Heintz (2009), with this type of sensory–motor learning (i.e., kinesthetic strategies), a person can use his/her body’s “muscle memory” to remember the motor movements required to accomplish a specific goal. To the best of our knowledge, routine topographic procedures (observed in the present research) have not been considered in the literature. From a cognitive neuroscience perspective, Simmons and Barsalou (2003, p. 9) argue that “substantial evidence exists for topographic organization in feature maps, hanging from the visual system to the motor, somatosensory, and auditory systems. In these areas, feature maps are often laid out according to the physical structure of the world.” Thus, it is not surprising that students mobilized 14.81% of topographical cues during the learning and memorization processes. In regard to kinesthetic strategies, Davidson (2009) implies that the development of motor memory programming is essential to provide both automaticity and fluency, which, in turn, functions as feedback for memorization.
According to the literature, there is some controversy regarding the role of kinesthetic representation in memorization (Lehmann, 1997). One cannot ignore that solely kinesthetic cues lend themselves well to rote memorization. As Wöllner and Williamon (2007) observed, the lack of this type of sensorial modality may weaken a memorized performance. Lisboa et al. (2009) reported that in written recall, serial cuing was impaired by the absence of sensorimotor cues, particularly for basic PCs. The authors concluded that basic PCs operated as part of a serial chain of association that reminded the musician about what came next. Such results suggest that professional musicians are more aware of the necessity of sensorimotor cues for written recall tasks.
In Phase II, as previously mentioned, the four categories of PCs were roughly equally utilized by the students. Considering each incidence of the employed PCs and the academic level of the participants, it seems that the number of employed PCs depicted a bell-shape curve: the first-year students (U2a and U2b) employed a low number of PCs; PCs increase among the students who are slightly further in their studies; and they finally decrease among the most experienced students (U7 and U8). Nevertheless, one cannot ignore that the peculiarities of each piece and each student’s relative expertise level may also have been responsible for the diverse rates of incidence observed among the participants.
The main finding extracted from the Phase II results, as shown in Figure 7, is that seven of the nine students (U2 to U7) chose interpretative PCs, suggesting that, in this sample, training in PC protocols sensitized the students to interpretative cues. The participants’ statements stress that they were quite willing to adopt and were capable of using the protocol proposed by Chaffin and collaborators, as shown in the excerpts below:
[With the protocol,] I learned to classify; there are things that I was already sort of thinking about but in an indirect way. I was thinking without needing to be explicit … When I’m playing, I always try to think about the interpretation, phrasing, climaxes, voice leading, articulations—not basic stuff. There are passages that still worry me, I mean, whether it is going to work out given the difficulty because it is a very long and difficult piece. (U6)
[Having had] contact with the PCs was very important to me … I read some stuff; I see my way of learning, for example. As far as the Mozart [sonata], I think that fifty percent is always going to be related to the mechanical issues and also topographical issues … but there are passages that I do not need to think about playing anymore … I still find myself thinking about sections because I want to finish the ideas … I think that the greater the number of explicit elements I can have and the more I get help on how to practice, the better my practice will be … I believe that this [the protocol] is most important for the slow movements because the hand in such passages ends up losing its place and inertia sets in … if I end up losing my place, then the cues will be even more useful. (U2b)
These results are in line with those reported by Chaffin, Demos, et al. (2009), who observed that college students used more interpretative PCs than professionals and graduate school students. As far as this last category is concerned, structural and basic cues received the most markings. We must then reaffirm that the employment of PC protocols provided the students in this sample with a meaningful strategy, considering that they willingly acknowledged their newly acquired awareness of interpretative features as integral to their memorization processes.
Although the relationship between tempo and the employment of performance cues has merited little discussion in the literature, Chaffin, Demos, et al. (2009) found an association between interpretative cues and tempo manipulations. In the lab performances, the effects of expressive and interpretative cues on tempo were consistent with the performer’s goals for playing with exaggerated, normal, or minimal expression. Decreasing in tempo at these cues was more frequent in extra-expressive performances and less frequent in non-expressive performances. The authors concluded that, in terms of tempo, all the statistically reliable differences between performances were attributable to performance cues, and there were no reliable differences due to other musical properties (Lisboa et al., 2007). In the present study, we found negative correlations (r = −0.751) between all PCs and tempo manipulations, as observed in Figure 10.
We also found further correlations between stylistic characteristics and the use of some specific types of PCs. As shown in Figure 11, stylistic characteristics (i.e., the historical period) seemed to affect the nature of the PCs chosen for the students’ memorized performances. One can see similarities among the students who played pieces by Mozart, Haydn and Beethoven (Op. 2, no. 3), which are likely attributable to the use of expressive PCs (students U7 and U8) and basic PCs (students U4b and U2b). Students U6 and U4c also reflected this pairing tendency in the ways in which they used expressive and interpretive PCs. Considering that the two did not share a common instructor, their results could have been related to the repertoire, a late Beethoven sonata and Chopin’s Ballade, both in A-flat. Student U4d (Beethoven’s Sonata Op. 81) applied PCs in a way that resembled both groups in terms of his reliance on basic, interpretative and expressive cues. Student U2a, in contrast, distinguished himself by his parsimonious use of PCs to perform Chopin’s Scherzo Op. 31, no. 2 in B-flat. He claimed he relied on PCs only after he had reached m. 64. Until that measure, he seemed to rely on the process of associative chaining memory. In this case (which is quite different from the other cases shown in the Figure 11 dendrogram), we cannot ignore the difficulty of this piece for this student. Chaffin, Demos, et al. (2009) do briefly mention that the employment of PCs may be affected by the musical style, in the case of four professional musicians. In a short sample, in which the same pianist played two pieces, Debussy’s Clair de Lune and the Presto from Bach’s Italian Concerto, the authors argued that there were fewer reported PCs for Debussy because that piece is comparatively easier than the Bach piece. Chaffin, Demos, et al. (2009) also found that professional pianists used a comparable number of PCs when playing stylistically diverse works, such as Debussy’s Clair de Lune, Bach’s Prelude from Cello Suite no. 8, Stravinsky’s Ricercar and Brahm’s Sonata no. 2 for piano. Even though the stylistic differences are significant, the use of PCs was not shown to be affected by the piece’s style or genre. As shown in the discussion, these results are markedly different from the ones obtained in the present study. We speculate that the piece’s style may affect PC use for the memorized performance. Further studies with larger samples are necessary to clarify this relationship.
Finally, as stated by Lehmann et al. (2007), memory relies on previous knowledge and the meaningful encoding of material. In addition, musicians, and human beings in general, have diverse inclinations, display a wide variety of learning preferences and, last but not least, come from considerably divergent backgrounds and cultural milieus. Given musicians’ individual histories, one cannot ignore that the conscious use of explicit means of memorization requires additional effort and deliberation. The choice and the manipulation of strategies can reveal the musician’s level of expertise. This research shows the extent to which a group of undergraduates understood and manipulated a memorization protocol for the first time.
Final remarks
As far as we know, there is no study in the literature that reports on undergraduate students’ use of Chaffin’s PCs. Notably, one cannot affirm or conclude that these PCs have been retained or automatized by these students because their practices were neither observed nor recorded. The present research focused on the memorized performance to promote students’ awareness of parameters of expression (phrase contour, dynamics, timing, etc.), as demonstrated by their manipulation of interpretative PCs.
In the first phase of this investigation, we observed one important aspect that was frequently mentioned by the participants: the arrangement of the piano keyboard and its influence on learning and memorization processes. Some students (four out of nine) made repeated assertions about the keyboard’s topography in close association with hand location and register. They also mentioned pattern configurations, types of movements and directions of movements. The repeated mention of the keyboard’s topography in association with movement and register led us to hypothesize that this way of thinking may have contributed to the students cementing their recently acquired knowledge of musical structure and resolutions that are related to the musical instrument being played. Nevertheless, these are instances of procedural thinking. It is also relevant to differentiate the keyboard’s topography from kinesthetic cues because the students’ comments associated musical configurations, or sound patterns, with fingering and actions on the keyboard. Should we then establish a new category of performance cues, one that combines musical structure with planned action? Until now, studies have been closely associated with the elements found in scores; our sample, however, pushed us to see how cues are applied to resolutions that are instrumental in nature. One obvious explanation is that students are not professionals; they do not play extensive repertoires, and their experience is far more limited. The students showed strong preferences for specific placements on the keyboard, which they referred to as “topography.” Recent cognitive neuroscience research helps us contextualize this proposition. During Phase II, this category was far less influential, and the novelty of a structured approach absorbed their complete attention.
The present findings may provide some implications for musical education, namely, (i) the notion of instrumental topography as directly related to procedural knowledge in piano learning, and (ii) the potential of using the PC protocol as an instrumental practice guide. Concerning the former, instrumental teaching would benefit from the attention paid to the specificities of musical procedural knowledge because some students may have musical procedures that relate to the keyboard’s topography, that is, a kind of procedure that forms a very close association with hand placement and register. The reliance on topography as a procedure may support teaching strategies and improve musical learning. It is our contention that the instrument’s topographical features may establish stronger relationships with musical patterns and configurations. In addition, the implementation of the PC protocol as part of the practice routine may lead to reflective practice habits that, in turn, will produce more mature and artistic performances.
In sum, the present research suggests two aspects that demand further investigation: (i) the role of topographical cues by students in different phases of the learning process and (ii) the relationship between PC use and composition style. In addition, we propose that studies combining procedural and declarative knowledge on the memorization processes can shed further light on this fascinating and crucial subject.
Funding
Cristina C. Gerling gratefully acknowledges CNPq’s grant. Regina Antunes Teixeira Dos Santos thanks CNPq for their financial support (CNPq 472652/2012–5).
References
|
Aiello, R., Williamon, A. (2002). Memory. In Parncutt, R., McPherson, G. E. (Eds.), The science and psychology of music performance (pp. 167–181). New York, NY: Oxford University Press. Google Scholar | Crossref | |
|
Chaffin, R. (2011). Thinking about performance: Memory, attention, and practice. In Williamon, A., Edwards, D., Bartel, L. (Eds.), Proceedings of the international symposium on performance science (pp. 689–699). Utrecht, The Netherlands: AEC. Google Scholar | |
|
Chaffin, R., Demos, A., Crawford, M. (2009). The PC-survey: How does use of performance cues vary across musicians, instruments, musical styles, and performance? In Williamon, A., Pretty, S., Buck, R. (Eds.), Proceedings of the international symposium on performance science (pp. 57–62). Auckland, New Zealand: AEC. Google Scholar | |
|
Chaffin, R., Gerling, C. C., Demos, A., Melms, A. (2013). Theory and practice: A case study of how Schenkerian analysis shaped the learning of Chopin’s Barcarolle. In Williamon, A., Goebl, W. (Eds.), Proceedings of the international symposium on performance science (pp. 21–26). Vienna, Austria: AEC. Google Scholar | |
|
Chaffin, R., Imreh, G. (1997). ‘Pulling teeth and torture’: Musical memory and problem solving. Thinking and reasoning, 3, 315–336. Google Scholar | Crossref | |
|
Chaffin, R., Imreh, G. (2002). Practicing perfection: Piano performance as expert memory. Psychological Science, 13, 342–349. Google Scholar | SAGE Journals | ISI | |
|
Chaffin, R., Imreh, G., Lemieux, A. F., Chen, C. (2003). ‘Seeing the big picture’: Piano practice as expert problem solving. Music Perception, 20, 465–490. Google Scholar | Crossref | ISI | |
|
Chaffin, R., Lemieux, A. F., Chen, C. (2007). ‘It’s different every time I play’: Spontaneity in highly prepared musical performance. Music Perception, 24, 455–472. Google Scholar | Crossref | ISI | |
|
Chaffin, R., Lisboa, T., Logan, T., Begosh, K. T. (2010). Preparing for memorized cello performance: The role of performance cues. Psychology of Music, 38, 3–30. Google Scholar | SAGE Journals | ISI | |
|
Chaffin, R., Logan, T. R., Begosh, K. T. (2009). Performing from memory. In Hallam, S., Cross, I., Thaut, M. (Eds.), The Oxford handbook of music psychology (pp. 364–376). New York, NY: Oxford University Press. Google Scholar | |
|
Davidson, J. W. (2009). Movement and collaboration in musical performance. In Hallam, S., Cross, I., Thaut, M. (Eds.), The Oxford handbook of music psychology (pp. 352–363). New York, NY: Oxford University Press. Google Scholar | |
|
Ebert, A., Deller, M., Steffen, D., Heintz, M. (2009). ‘Where did I put that?’: Effectiveness of kinesthetic memory in immersive virtual environments. In Stephanidis, C. (Ed.), Universal access in human–computer interaction (pp. 178–188). Heidelberg, Germany: Springer. Google Scholar | Crossref | |
|
Ericsson, K. A. (2008). Deliberate practice and acquisition of expert performance: A general overview. Academic Emergency Medicine, 15, 988–994. Google Scholar | Crossref | Medline | ISI | |
|
Ericsson, K. A., Kintsch, W. (1995). Long-term working memory. Psychological Review, 102, 211–245. Google Scholar | Crossref | Medline | ISI | |
|
Finney, S. A., Palmer, C. (2003). Auditory feedback and memory for music performance: Sound evidence for an encoding effect. Memory and Cognition, 31, 51–64. Google Scholar | Crossref | Medline | ISI | |
|
Gerling, C. C., Dos Santos, R. A. T. (2011). An evaluation of parameters in performance: The effects of aural and visual stimuli. In Williamon, A., Edwards, D., Bartel, L. (Eds.), Proceedings of ISPS 2011 (pp. 509–514). Utrecht, The Netherlands: AEC. Google Scholar | |
|
Hair, J. F., Black, W. C., Babin, B. B., Anderson, R. E. (2013). Multivariate data analysis (7th ed.). Harlow, UK: Pearson. Google Scholar | |
|
Hallam, S. (1997). Approaches to instrumental music practice of experts and novices: Implications for education. In Jørgensen, H., Lehmann, A. C. (Eds.), Does practice make perfect? (pp. 89–107). Oslo, Norway: Norges musikkhøgskole. Google Scholar | |
|
Krampe, R. T., Ericsson, K. A. (1996). Maintaining excellence: Deliberate practice and elite performance in young and older pianists. Journal of Experimental Psychology: General, 125, 331–359. Google Scholar | Crossref | Medline | ISI | |
|
Lehmann, A. C. (1997). Acquired mental representations in music performance: Anecdotal and preliminary empirical evidence. In Jørgensen, H., Lehmann, A. C. (Eds.), Does practice make perfect? (pp. 141–163). Oslo, Norway: Norges musikkhøgskole. Google Scholar | |
|
Lehmann, A. C., Ericsson, K. A. (1997). Expert pianists’ mental representations: Evidence from successful adaptation to unexpected performance demands. In Gabrielsson, A. (Ed.), Proceedings of the third triennial ESCOM conference (pp. 165–169). Uppsala, Sweden: Uppsala University. Google Scholar | |
|
Lehmann, A. C., Sloboda, J. A., Woody, R. H. (2007). Psychology for musicians: Understanding and acquiring skills. Oxford, UK: Oxford University Press. Google Scholar | Crossref | |
|
Lisboa, T., Chaffin, R., Demos, A. (2013). Recording thoughts as an aid to memorization: A case study. In Williamon, A., Goebl, W. (Eds.), Proceedings of the international symposium on performance science (pp. 625–630). Vienna, Austria: AEC. Google Scholar | |
|
Lisboa, T., Chaffin, R., Logan, T. (2009). How memory fades: Very-long-term recall of Bach. In Williamon, A., Edwards, D., Bartel, L. (Eds.), Proceedings of the international symposium on performance science (pp. 315–320). Utrecht, The Netherlands: AEC. Google Scholar | |
|
Lisboa, T., Chaffin, R., Logan, T. (2011). A self-study of practice: Words versus action in music problem solving. In Williamon, A., Edwards, D., Bartel, L. (Eds.), Proceedings of the international symposium on performance science 2011 (pp. 517–522). Utrecht, The Netherlands: European Association of Conservatoires (AEC). Google Scholar | |
|
Lisboa, T., Chaffin, R., Logan, T., Begosh, K. (2007). Variability and automaticity in highly practiced cello performance. In Williamon, A., Coimbra, D. (Eds.), Proceedings of the international symposium on performance science (pp. 160–165). Utrecht, The Netherlands: AEC. Google Scholar | |
|
Noice, H., Jeffrey, J., Noice, T., Chaffin, R. (2008). Memorization by a jazz musician: A case study. Psychology of Music, 36, 63–79. Google Scholar | SAGE Journals | ISI | |
|
Simmons, K., Barsalou, L. W. (2003). The similarity in topography principle: Reconciling theories of conceptual deficits. In The organization of conceptual knowledge in the brain: Neuropsychological and neuroimaging perspective (pp. 451–486). Hove, UK: Psychology Press. Google Scholar | Crossref | |
|
Smith, J. A., Osborn, M. (2003). Interpretative phenomenological analysis. In Smith, J. A. (Ed.), Qualitative psychology: A practical guide to methods. London, UK: Sage. Google Scholar | |
|
Williamon, A., Egner, T. (2004). Memory structures for encoding and retrieving a piece of music: An ERP investigation. Cognitive Brain Research, 22, 36–44. Google Scholar | Crossref | Medline | |
|
Wöllner, C., Williamon, A. (2007). An exploratory study of the role of performance feedback and musical imagery in piano playing. Research Studies in Music Education, 29, 39–54. Google Scholar | SAGE Journals |