Sydney Scott
Poster Number
10a
Introduction/Abstract
Over the past three decades, chunking practice has become increasingly popular for refining fast passagework in intermediate and advanced string studios across the United States. So far however, only limited research has been done testing the efficacy of chunking practice within string pedagogy. Nonetheless, there is much to be learned about chunking from prior studies in psychology, neurology, and exercise physiology on sequencing, motor learning, and memory in music. In the following article we will explore both theoretical and implied effects of chunking practice for beginning and advanced string students. Chunking, also known as sequencing, grouping, or practicing in parts, can be used to learn a variety of tasks from memorizing a phone number to learning the letters of the alphabet. Chunking can be used in general music classes to teach students new songs by rote using line-by-line echoing. In the case of instrumental practice, chunking typically consists of fast, repetitive, note sequences with a specific focus. Chunking practice is commonly used in string pedagogy to teach fast, technically demanding passages. Opinions vary as to the ideal tempo and focuses for implementation of chunking practice. For the sake of consistency throughout this review, chunking practice will refer to repeated groups of 2-16 notes played at goal tempo (unless otherwise stated). It is important to note that the efficacy of chunking isn’t necessarily consistent across subjects; however, much can be learned from prior research in other domains. Psychologist K. Lashley (1951) was the first to propose the concept known as motor chunking.[1] Independent but related to chunking practice, motor chunking is the cognitive process of long term memory development through the motor hierarchy in order to trigger a more precise and direct execution process. The more direct the execution process, the more room available for individual musical interpretations and nuances (Palmer, 2000). Rosenbaum (1986) further explores this concept in his paper on Hierarchical Control of Rapid Movement Sequences, suggesting that simple movements chunked together as one response can, in fact, be received faster and more accurately than individual, independently triggered movements. This theoretical realization and its assumed potential, is in large part, why chunking has grow in popularity over the past few decades. For over a century, scholars in a variety of fields have debated the superiority of a whole vs. parts approach to learning. In 1900s, Steffen published initial findings on progressive-part vs. whole learning methods finding a whole method approach overall superior. Soon after however, Meumann, Reed, and Pechstein, conducted a similar study on whole vs. part motor control learning yielding directly conflicting results (as cited in Shay, 1934). Since, numerous conflicting studies on whole vs. part learning methods have been conducted (Shay, 1934). Although the debate continues, several fields have developed specific criteria for determining the most effective practice approach in any given situation. Sports science scholars for example, recognize the importance of addressing complexity of movements in relationship between each part to the greater whole (Fontana, Furtadao, Mazzardo, & Gallagher, 2009). In doing so, they have established guidelines for the most efficient motor learning approaches based of complexity and organization of tasks. According to Naylor and Briggs (1963) guidelines, complex and un-organized tasks (tasks for which one step doesn’t influence the execution of the next) are more effectively practiced in parts; where as, less complex and organized tasks are more effectively learned as a whole. Transferring these criteria to the realm of musical practice, one can postulate chunking practice to be most effective (if at all) when learning fast, unpredictable passages. Effective practice of fast passagework on a string instrument consists of the progressive addition and refinement of various musical elements. The specific order in which each element is added can influence practice success due to the motor learning hierarchy. The motor learning hierarchy is the sequence order and in which tasks are most effectively learned and stored into long-term memory. In order to play each note, a musician must continuously cycle between a phase of selection and execution. The initial hierarchical phase of selection, in this case, consists of processing both how the left finger(s) and right arm need to move in order to play the next note as intended. This initial phase is the most easily adaptable with practice and is consistent with music performance studies recognizing largest improvements in skill during initial stages (Palmer & Drake, 1997). Through explicit memory recall processing of task instruction, once stored as a long-term memory, the selection phase can directly elicit the execution phase to begin at the recognition of initial prompts (Diedrichsen & Kornvsheva, 2015). Although slightly less adaptable, the execution phase, or physical finger placement and bow arm movement response, can also be shortened through musical practice after refinement of the initial selection process (Palmer, 2000). Although only a very minute time difference, the efficiency and speed of each finger placement (and lift) becomes quite significant in overall tempo and execution. Superior development of finger Figure 1. Levels of skill learning (Diedrichsen & Kornysheva) dexterity and speed is required to accurately play many standard violin concertos and audition excerpts. Even the eyes must learn to adapt to meet faster tracking and focusing requirements by forming additional visual motor primitives (Diedrichsen & Kornysheva, 2015). Although both selection and execution tasks are performed cyclically, it is important to take note that, due to the hierarchical nature of this model, the neurologic components involved in the execution phases don’t appear to begin developing until selection skills have, at minimum, established a solid base (Diedrichsen & Kornysheva, 2015). As a result, one can infer a variance between ideal practice strategies for beginning and advanced string students. Young students tend to use a limited number of practice strategies, most commonly repetition (Sikes, 2013). Although running a piece of music is commonly considered most helpful as a way to initially introduce and then finally polish repertoire, it can still be a very effective practice method, especially for beginning players who still spend the majority of their time in the task selection phase. In observing beginning 3rd and 4th grade instrumentalists, McPherson (2005) found students able to “apply musically appropriate mental strategies very early on in their learning were more likely to succeed in comparison with their peers.” These students have not yet developed an efficient process for finger selection and execution and are not able to apply motor chunking strategies. Nonetheless, even when continuously repeating a piece of music at the same speed, with the right mental strategies, beginning students are able to improve incredibly rapidly. Although variation of strategies may not be as crucial for beginning students, advanced string players show the understanding and frequent utilization of a wide variety of practice methods. Although there does not seem to be a standard set of approaches for all advanced musicians, professional musicians do appear to be very systematic when selecting their practice approach for any given technical challenge (Sikes, 2013). Not surprisingly, during the process of mastering a piece, advanced musicians constantly continue adjusting their approach to best meet their musical goals (Chaffin et al., 2010; Chaffin & Imreh, 2001; Nielsen, 1999). Interestingly, at all ages and experience levels, free practice tends to consistently prove to be at least equally effective, if not more, than any specific implemented approaches (Sikes, 2013). This may seem surprising considering how much we value introducing specific practice styles to our students. However, once acquiring a wide variety of practice approaches and knowing when to apply them, a variety of technical challenges can immediately be addressed with the most efficient approach. Certain specific practice approaches naturally allow for more variance than others. In his doctoral dissertation Donald (1977) took a unique approach to varied practice with intermediate and advanced piano students. Donald investigated the effects of two different varied practice approaches in which students were asked to practice a scale either at a slow tempo, faster tempo, and then goal tempo, or by consistently altering between a slow and goal tempo until their target tempo was reached (Donald, 1977). Surprisingly, those in the second group reached their target tempo significantly faster (Donald, 1997). A similar study performed by Paul Henley (2001), tested high school woodwind players before and after a total of seven performances of full piece comparing use of slow to fast, at tempo, and alternating slow to fast practice techniques. The results in Henley’s study however, showed no significant variation of technique or tempo between players (Henley, 2001). By having students approach their practice by solely running the entire piece regardless of the tempo, participants were unable to focus in on the hardest sections of the work. The most recent study comparing practice approaches for fast passagework was completed by Sikes (2013). Sikes compared the success rates of university string players’ when practicing using a free, slow to fast, chunking, and repetition practice approach. With a similar timeline (same day pre and post test) to that of Donald and Henley’s studies, Sikes found the free practice group to have the most success followed by the chunking group (Sikes 2013). One of the most noticeable differences between Sikes study and that of both Donald and Henley, is Sikes initially gave students one minute to look over and silently practice the music before the pre test (Donald, 1997; Henley, 2001; Sikes, 2013). One may think this is irrelevant as it was allowed prior to the pretest and each group was designated this same one minute of mental practice. Nonetheless, the additional musical foresight gained from silent practice before beginning designated practice, could have allowed more advanced students in the chunking group an advantage. By already refining the selection phase through initial free practice, chunking would have a greater likelihood of influencing efficiency of execution through motor chunking. Not all 40 individual’s results are currently accessible, it may however be beneficial to compare the results of the most and least advanced students within the chunking group to test this theory. One can gain considerable additional insights on previous data when carefully analyzed from a new perspective. The compilation and analysis of previous research on chunking efficacy along with insights from research in exercise physiology and neurology, was very beneficial in constructing a new approach for further studies.
Purpose
This study will test the effects of chunking vs. slow to fast practice techniques in both beginning and advanced undergraduate violin players. The three main objectives of this study are 1) to determine whether using the ‘chunking’ or slow to fast practice method is a more effective for beginning string technique players learning to play fast passages 2) to determine whether a chunking or slow to fast practice method is more effective for advanced string players learning to play fast passages and 3) to determine whether or not the efficacy of each practice model varies proportionally between both subject groups based on experience level. Although applicable to a variety of repertoire, chunking appears to have the most potential for fast, in-consistent passagework. This pilot study was designed to test the effects of a chunking practice model for fast passagework performed by both beginning and advanced string students. In order to test the efficacy of chunking practice for fast passagework, results were be compared alongside those of the standard slow to fast practice model.
Method
The efficacy of each practice model was tested for both beginning (group 1) and advanced (group 2) students. Beginning participants, university students from string technique class at University of the Pacific, each had 1-12 months prior playing experience on the violin before completing the study. Advanced participants, university students from the violin studio at University of the Pacific, all had over five years experience playing violin. Participants were assigned to either the slow to fast or chunking group using stratified randomization. The chunking group (group B), was assigned to begin with small groupings of 2-4 notes and gradually increase length of groupings over time. The chunking group was also specifically assigned to overlap note groupings veering away from influence of measure lines. Note groupings were to consistently be played at goal tempo throughout practice sessions. Those in the slow to fast practice group (group A) were assigned to practice the given etude all the way through at a comfortable tempo gradually increasing tempo as comfortable. Each subject participated in an audio recorded pre and posttest as well as five additional independent practice sessions over the course of a week. Participants were asked to devote 55 minutes of their time to the study consisting of a 15-minute pre test and meeting, five independent practice sessions, and a 15-minute posttest and meeting. Each 15-minute pre meeting consisted of five minutes of free practice, a pre test, and five minutes of guided practice using assigned practice technique. Posttest meetings consisted of five minutes of assigned focus practice, posttest, overview of study, and introduction of additional practice technique. For pre and posttests, participants were given two chances to run etude all the way through at a tempo independently selected by participant. Both pre and posttests were audio recorded by proctor using a Voice Recorder application. For participants’ comfort, proctor left the room during playing tests. Goal tempos were assigned based on participants’ skill level and/or initial pre test performance tempo. Beginning students were assigned a goal tempo for etude of their initial pre test tempo (in quarter notes) plus 50 beats per minute. Advanced students were assigned the goal tempo of 144 beats per minute as inspired by tempo of standard violin audition repertoire. All participants used a metronome consistently throughout pre/post tests and all practice sessions. Primary researcher composed etudes for this study for both beginning and advanced violin students. The etude for the beginning group (group 1) consisted of eight measures of continuous eight notes using various combinations of A, B, C#, and D on the A string (L. fingers: none, pointer, middle, and ring). For advanced participants, the etude consisted of 11 measures of consistent 16th notes using various combinations of A, B, C#, D, and E on the A string (L. fingers: none, pointer, middle, ring, and pinky). Both etudes were designed to be relatively simple technically in order for highest potential success when played at a faster tempo. Nonetheless, etudes were also intentionally written with an unpredictable melodic contour, as it is inferred chunking has the highest potential for success relative to other approaches when studying a fast and unpredictable tempo. Dr. Ann Miller, D.M.A. Program Director for Strings at University of the Pacific, generously aided in etude compositions to ensure appropriate skill and experience level for subjects.
Results
Results indicated that out of the two beginning groups, those following the slow to fast practice method yielded a significantly larger average tempo percentage gain of all participants. Out of the two advanced groups, those following the chunking method yielded a significantly higher average tempo percentage gain than those following the slow to fast practice method. In attempt to account for participants who were unable to complete all five practice sessions over the course of the week (two participants), daily improvement averages were taken and added to initial results in order estimate additional improvement assuming linear growth. (y=5x/n +t, y=adjusted averages, x=initial gain, n=number of days practiced, t=initial pre test tempo) There appears to be a wider discrepancy in improvement between individuals in both the beginning and advanced chunking group than that of the slow to fast group. Observing the difference between initial and end tempos for the slow to fast practice groups (1A and 2A) and comparing it to the tempo differences for of the chunking group (1B and 2B), one can observe a considerably larger degree of variance in the chunking groups than that of the slow to fast groups.
Significance
Comparison of pre and post tempo variance suggests that a gradual progression from slow to fast practice is more beneficial to beginning students than chunking practice. Results also suggest that chunking practice is more beneficial, overall, for advanced than beginning violinists. However, the larger variance in initial and ending tempo difference between individuals, suggests certain individuals benefit more from the chunking approach than others. Observing participants’ level of success while using chunking practice based on prior experience with practice technique, opinion of method, anxiety level, etc. may all aid in further explanations behind variance between participant results. An important observation to account for is some players are more cautious in nature than others. Based on their personality, some players may be more inclined to set a slower tempo than they are capable of playing at. Luckily, these variations usually balance out when comparing pre and posttests, however we cannot assume participants to be in the same mental state during both tests. Students not able to complete all practices, in return, had differing results. In attempt to consider this deficit when interpreting results, an average of results was taken and extended to estimate additional improvements from further practice. As known from individual experience, practice is rarely linear thus making this approach unreliable. Also, in assigning participants to choose their own tempo for both the pre and posttest, metronome tempos were not always reflective of student’s playing levels and abilities. Although note errors and number of stops were recorded, they were not represented in the overall outcome of the study and tempo neither consistency nor tone quality were explicitly accounted for. Sometimes participants were more inclined to speed through a piece and choose to repeat their second play through at the same tempo regardless of any musical elements that may be hampered. An extended version of this pilot study allowing participants assigned the chunking method additional time for initial slow practice (to allow time for development of selection accuracy and efficiency) would further prove or disprove the effects of chunking in certain subject groups. A similar extended study would also have the potential to show how the long-term learning curve of one method relates to the other. When using the slow to fast practice method, string players often struggle to get over a certain tempo around 132-136 bpm that can take weeks or months to overcome before reaching a final 144 bpm performance tempo. It would be interesting to observe if any similar distinct lulls in improvement occur regularly for students using the chunking approach.
Location
DeRosa University Center
Format
Poster Presentation
Poster Session
Morning 10am-12pm
Sydney Scott
DeRosa University Center
Over the past three decades, chunking practice has become increasingly popular for refining fast passagework in intermediate and advanced string studios across the United States. So far however, only limited research has been done testing the efficacy of chunking practice within string pedagogy. Nonetheless, there is much to be learned about chunking from prior studies in psychology, neurology, and exercise physiology on sequencing, motor learning, and memory in music. In the following article we will explore both theoretical and implied effects of chunking practice for beginning and advanced string students. Chunking, also known as sequencing, grouping, or practicing in parts, can be used to learn a variety of tasks from memorizing a phone number to learning the letters of the alphabet. Chunking can be used in general music classes to teach students new songs by rote using line-by-line echoing. In the case of instrumental practice, chunking typically consists of fast, repetitive, note sequences with a specific focus. Chunking practice is commonly used in string pedagogy to teach fast, technically demanding passages. Opinions vary as to the ideal tempo and focuses for implementation of chunking practice. For the sake of consistency throughout this review, chunking practice will refer to repeated groups of 2-16 notes played at goal tempo (unless otherwise stated). It is important to note that the efficacy of chunking isn’t necessarily consistent across subjects; however, much can be learned from prior research in other domains. Psychologist K. Lashley (1951) was the first to propose the concept known as motor chunking.[1] Independent but related to chunking practice, motor chunking is the cognitive process of long term memory development through the motor hierarchy in order to trigger a more precise and direct execution process. The more direct the execution process, the more room available for individual musical interpretations and nuances (Palmer, 2000). Rosenbaum (1986) further explores this concept in his paper on Hierarchical Control of Rapid Movement Sequences, suggesting that simple movements chunked together as one response can, in fact, be received faster and more accurately than individual, independently triggered movements. This theoretical realization and its assumed potential, is in large part, why chunking has grow in popularity over the past few decades. For over a century, scholars in a variety of fields have debated the superiority of a whole vs. parts approach to learning. In 1900s, Steffen published initial findings on progressive-part vs. whole learning methods finding a whole method approach overall superior. Soon after however, Meumann, Reed, and Pechstein, conducted a similar study on whole vs. part motor control learning yielding directly conflicting results (as cited in Shay, 1934). Since, numerous conflicting studies on whole vs. part learning methods have been conducted (Shay, 1934). Although the debate continues, several fields have developed specific criteria for determining the most effective practice approach in any given situation. Sports science scholars for example, recognize the importance of addressing complexity of movements in relationship between each part to the greater whole (Fontana, Furtadao, Mazzardo, & Gallagher, 2009). In doing so, they have established guidelines for the most efficient motor learning approaches based of complexity and organization of tasks. According to Naylor and Briggs (1963) guidelines, complex and un-organized tasks (tasks for which one step doesn’t influence the execution of the next) are more effectively practiced in parts; where as, less complex and organized tasks are more effectively learned as a whole. Transferring these criteria to the realm of musical practice, one can postulate chunking practice to be most effective (if at all) when learning fast, unpredictable passages. Effective practice of fast passagework on a string instrument consists of the progressive addition and refinement of various musical elements. The specific order in which each element is added can influence practice success due to the motor learning hierarchy. The motor learning hierarchy is the sequence order and in which tasks are most effectively learned and stored into long-term memory. In order to play each note, a musician must continuously cycle between a phase of selection and execution. The initial hierarchical phase of selection, in this case, consists of processing both how the left finger(s) and right arm need to move in order to play the next note as intended. This initial phase is the most easily adaptable with practice and is consistent with music performance studies recognizing largest improvements in skill during initial stages (Palmer & Drake, 1997). Through explicit memory recall processing of task instruction, once stored as a long-term memory, the selection phase can directly elicit the execution phase to begin at the recognition of initial prompts (Diedrichsen & Kornvsheva, 2015). Although slightly less adaptable, the execution phase, or physical finger placement and bow arm movement response, can also be shortened through musical practice after refinement of the initial selection process (Palmer, 2000). Although only a very minute time difference, the efficiency and speed of each finger placement (and lift) becomes quite significant in overall tempo and execution. Superior development of finger Figure 1. Levels of skill learning (Diedrichsen & Kornysheva) dexterity and speed is required to accurately play many standard violin concertos and audition excerpts. Even the eyes must learn to adapt to meet faster tracking and focusing requirements by forming additional visual motor primitives (Diedrichsen & Kornysheva, 2015). Although both selection and execution tasks are performed cyclically, it is important to take note that, due to the hierarchical nature of this model, the neurologic components involved in the execution phases don’t appear to begin developing until selection skills have, at minimum, established a solid base (Diedrichsen & Kornysheva, 2015). As a result, one can infer a variance between ideal practice strategies for beginning and advanced string students. Young students tend to use a limited number of practice strategies, most commonly repetition (Sikes, 2013). Although running a piece of music is commonly considered most helpful as a way to initially introduce and then finally polish repertoire, it can still be a very effective practice method, especially for beginning players who still spend the majority of their time in the task selection phase. In observing beginning 3rd and 4th grade instrumentalists, McPherson (2005) found students able to “apply musically appropriate mental strategies very early on in their learning were more likely to succeed in comparison with their peers.” These students have not yet developed an efficient process for finger selection and execution and are not able to apply motor chunking strategies. Nonetheless, even when continuously repeating a piece of music at the same speed, with the right mental strategies, beginning students are able to improve incredibly rapidly. Although variation of strategies may not be as crucial for beginning students, advanced string players show the understanding and frequent utilization of a wide variety of practice methods. Although there does not seem to be a standard set of approaches for all advanced musicians, professional musicians do appear to be very systematic when selecting their practice approach for any given technical challenge (Sikes, 2013). Not surprisingly, during the process of mastering a piece, advanced musicians constantly continue adjusting their approach to best meet their musical goals (Chaffin et al., 2010; Chaffin & Imreh, 2001; Nielsen, 1999). Interestingly, at all ages and experience levels, free practice tends to consistently prove to be at least equally effective, if not more, than any specific implemented approaches (Sikes, 2013). This may seem surprising considering how much we value introducing specific practice styles to our students. However, once acquiring a wide variety of practice approaches and knowing when to apply them, a variety of technical challenges can immediately be addressed with the most efficient approach. Certain specific practice approaches naturally allow for more variance than others. In his doctoral dissertation Donald (1977) took a unique approach to varied practice with intermediate and advanced piano students. Donald investigated the effects of two different varied practice approaches in which students were asked to practice a scale either at a slow tempo, faster tempo, and then goal tempo, or by consistently altering between a slow and goal tempo until their target tempo was reached (Donald, 1977). Surprisingly, those in the second group reached their target tempo significantly faster (Donald, 1997). A similar study performed by Paul Henley (2001), tested high school woodwind players before and after a total of seven performances of full piece comparing use of slow to fast, at tempo, and alternating slow to fast practice techniques. The results in Henley’s study however, showed no significant variation of technique or tempo between players (Henley, 2001). By having students approach their practice by solely running the entire piece regardless of the tempo, participants were unable to focus in on the hardest sections of the work. The most recent study comparing practice approaches for fast passagework was completed by Sikes (2013). Sikes compared the success rates of university string players’ when practicing using a free, slow to fast, chunking, and repetition practice approach. With a similar timeline (same day pre and post test) to that of Donald and Henley’s studies, Sikes found the free practice group to have the most success followed by the chunking group (Sikes 2013). One of the most noticeable differences between Sikes study and that of both Donald and Henley, is Sikes initially gave students one minute to look over and silently practice the music before the pre test (Donald, 1997; Henley, 2001; Sikes, 2013). One may think this is irrelevant as it was allowed prior to the pretest and each group was designated this same one minute of mental practice. Nonetheless, the additional musical foresight gained from silent practice before beginning designated practice, could have allowed more advanced students in the chunking group an advantage. By already refining the selection phase through initial free practice, chunking would have a greater likelihood of influencing efficiency of execution through motor chunking. Not all 40 individual’s results are currently accessible, it may however be beneficial to compare the results of the most and least advanced students within the chunking group to test this theory. One can gain considerable additional insights on previous data when carefully analyzed from a new perspective. The compilation and analysis of previous research on chunking efficacy along with insights from research in exercise physiology and neurology, was very beneficial in constructing a new approach for further studies.
