Abstract
The purpose of this study was to investigate the effects of internal versus external focus of attention during novice teachers’ instructional preparation on their subsequent rehearsal behaviors. Thirty-two undergraduate instrumental music education students led bands in a series of three, 6-minute rehearsals on their assigned excerpt. Prior to these rehearsals, participants engaged in condition-specific score study and rehearsal preparation activities. Internal group (n = 16) participants’ preparation related to knowledge of the score, whereas external group (n = 16) participants focused their preparations on observable rehearsal behaviors with a minimal amount of time devoted to score study. We systematically analyzed video recordings of these rehearsals, calculating rates per minute of teacher verbalizations across several performance and teaching variables. We found that compared to the internal group, the external group exhibited higher rates of positive/specific feedback, conducted more frequent and briefer performance segments, and more often asked for the ensemble to start without providing a directive. The internal group mentioned ensemble balance in their rehearsals more frequently than did the external group, and their verbalizations reflected greater concern for Tone Quality. A panel of independent evaluators viewed all 96 video recordings of the rehearsals (presented to them without sound) and rated the clarity and the expression of participants’ conducting. We found a significant time by condition interaction for Expression, with scores for the external group increasing over time and scores for the internal group decreasing. We suggest that these results reflect the distinct and complementary benefits of each of these preparation methods for novice music teachers.
The manner in which one prepares for a given task might presumably be reflected in one’s performance of that task. This would seem to hold true regarding a dynamic and complex undertaking such as leading music ensemble rehearsals—especially for novices, for whom requisite musical and instructional skills are still developing. Although undergraduate music education majors often have extensive experience as successful ensemble members, preparing to assume leadership of an ensemble requires an expanded repertoire of knowledge and skills and the application of extant musical and interpersonal skills in new contexts. Music teacher education courses commonly focus on both score study and rehearsal techniques, but professional opinions differ regarding when and how component skills within these areas should be learned by preservice music teachers (Buell, 1990; Ellis, 1994; Feldman, Contzius, & Lutch, 2011; Manfredo, 2008; Romines, 2003; Williamson, 1998). Empirical inquiry regarding these matters has been an area of ongoing interest to researchers (Lane, 2006; Silvey, 2011; Silvey & Montemayor, 2014).
Following different protocols for rehearsal preparation would logically yield concomitantly different results in both the evaluation of rehearsal leadership and in the analysis of discrete rehearsal behaviors. Still, the relationship between the perception of music teaching and the behavioral components of music instruction is not always clear (MacLeod & Nápoles, in press; Montemayor, 2014; Nápoles & MacLeod, 2013). Dichotomous comparisons of most preferred and least preferred teaching episodes have revealed empirical differences in the pace and/or content of student and teacher activity (Duke, Prickett, & Jellison, 1998; Yarbrough & Madsen, 1998). However, instructional episodes that are markedly different in terms of the frequency of negative teacher feedback may be perceived as similar to one another by both students (Duke & Henninger, 1998) and undergraduate observers (Duke & Henninger, 2002). Evaluations of teaching sometimes appear to be influenced by observer focus of observation (Duke & Prickett, 1987; Yarbrough & Henley, 1999) or by requiring observers to record specific aspects of teacher behavior (Duke & Blackman, 1991). Accordingly, it is possible that when comparing teaching activity between distinct experimental groups, behavioral differences may exist even when evaluations of teaching are found to be similar. This may be especially true for investigations in which the effects of a model on musical behavior are among the variables of interest; previous studies have generally shown such effects to be small in individuals’ performance behaviors or evaluations (Cash, Allen, Simmons, & Duke, 2014; Hewitt, 2001), ensemble performance evaluations (Morrison, 2002; Morrison, Montemayor, & Wiltshire, 2004), and novices’ teaching behaviors (Montemayor & Moss, 2009).
The purpose of this study was to compare effects of two different preparation methodologies—one centered around typical score study procedures and another focused more on observing and improving recognizable teacher behaviors—on novice teachers’ rehearsing and conducting. Inasmuch as possible, the experimental treatments for this investigation reflect a distinction between “what to teach” and “how to teach.” Regarding the latter, several research efforts have demonstrated positive effects of systematic self-review of music teaching behaviors (Johnston, 1993; Lethco, 1999; Worthy, 2005; Yarbrough, 1987). Many other studies have investigated aspects of score study as related to conducting expressivity (Treviño, 2008), ensemble member evaluations (Silvey, 2011), and error detection (Crowe, 1996; Hochkeppel, 1993; Hopkins, 1991; Van Oyen & Nierman, 1998). Certainly, both are important; the perceived emphasis on one at the expense of the other in music teacher preparation programs elicited passionate commentary from one prominent ensemble conductor:
We have too many people teaching music who don’t know music . . . their behavior is not artistic. I call them “musical mechanics.” . . . This attitude is a big problem, and much of this can be traced back to teacher preparation. The current product of many music degree programs is frightening. . . . They may know how to teach, but they don’t make music. . . . However, without that strong source of musical artistry, you have the elaborate structure of the Mississippi River, with all the levees, and the bridges and the parks along its banks . . . without any water! (Battisti, in Williamson, 1998, p. 7)
Nevertheless, existing theoretical models of expert teaching throughout the broader educational literature and in music education as well have long included aspects of subject-specific pedagogy as a special form of expertise, specific to subject matter but distinct from content knowledge (Berliner, 1986; Shulman, 1986, 1987; Standley & Madsen, 1991). Baumert and colleagues (2010), in a yearlong study of secondary-level mathematics instruction, found that teachers’ pedagogical content knowledge (PCK) was a greater predictor of student achievement than was content knowledge (CK) itself. Absent sufficient PCK, subject matter expertise may even hinder teachers’ instructional approach (Nathan & Petrosino, 2003). To the extent that skills in ensemble leadership constitute a form of PCK for music teaching, investigating approaches to instructional preparation that distinguish CK from PCK would seem to be an important research endeavor.
To our knowledge, no researchers have attempted to directly compare musical preparation against instructional preparation among preservice teachers. The present report is an extension of our earlier research effort (Silvey & Montemayor, 2014), where we examined effects of these strategies on rehearsal evaluations over the course of three band rehearsals. There, we found no significant differences between conditions on any of several dependent measures, including participants’ self-evaluation of their teaching, participants’ evaluation of ensemble performance, ensemble members’ evaluations of conductor rehearsal effectiveness and of conductor score knowledge, and audio evaluation of ensemble performance. Self-evaluations and ensemble member evaluations improved significantly over time in both conditions. In an effort to further elucidate possible preparation-related effects on novices’ teaching, for the current study we chose to systematically observe participants’ rehearsal behaviors, using video recordings generated from our earlier project.
Null hypotheses for the current study were that there would be no differences among novice teachers’ (a) rates of verbalizations during rehearsals related to selected performance, teaching, and rehearsal variables 1 ; (b) distribution of verbalizations related to categorized performance variables; or (c) clarity and expression of their conducting when comparing participants who had focused their rehearsal preparations using typical score study methodologies versus those whose focus was on observable rehearsal behaviors. Using terminology adapted from research in motor behavior learning (see Wulf, 2013, for a review), and mindful of the analogies thereof to that literature, we labeled these approaches as an internal focus and an external focus, respectively.
Method
All data were derived from our previous, related project (Silvey & Montemayor, 2014), to which we refer readers regarding more extensive details of the experimental methodology for our current investigation. In summary, 32 undergraduate instrumental music education students led bands in a series of three 6-minute rehearsals on an assigned excerpt of grade 2 (of 6) level band music, matched between two experimental conditions. Bands were composed of undergraduate musicians playing either primary or secondary instruments who did not participate as conductors in this study. Prior to these rehearsals, we led participants in intensive score study and rehearsal preparation activities. Internal group (n = 16) participants’ preparation related to knowledge of the score and development of an aural image of the music, whereas external group (n = 16) participants focused their preparations on observable rehearsal behaviors with a minimal amount of time devoted to score study. Between rehearsals, conductors continued rehearsal preparation on their own using techniques similar to the ones they were taught during the instructional sessions. (We confirmed participants’ compliance in this matter in the original study.)
The three rehearsals were completed at weekly intervals. We instructed participants to begin the first 6-minute rehearsal with a sight-read “run-through” performance of their assigned excerpt (about 1.5 minutes in duration) and to end the second and third rehearsals with a similar performance. Participants were allowed to use the remaining rehearsal time (about 4.5 minutes in each episode) in any manner they wished, with the instruction to elicit as much positive change in the ensemble as possible so that the final “run-through” performance at the end of the third rehearsal matched their musical intentions for the piece. All 96 rehearsals (32 teachers [16 at each institution] × 3 rehearsals) were video recorded, with the camera trained on the conductor. Digital video files were created to facilitate systematic observation.
For the current investigation, we measured frequencies of teachers’ verbal behaviors across 16 musical performance categories and seven teaching variables, using procedures and operational definitions as found in an earlier study (Montemayor & Moss, 2009). These variables are listed in Table S1 (available online at http://jrme.sagepub.com/supplemental). We tallied an occurrence of each dependent variable when the category was mentioned by name or was implied by the context of the teacher’s verbalizations. Instructions implied through conducting gestures were not counted. For any given rehearsal episode, any particular performance category was tallied only once per section within the music or ensemble. For example, upon stopping the ensemble, a conductor’s protracted explanation of dynamics was counted as one “dynamics” verbalization, regardless of how many times the word itself was used or the concept was addressed, unless the explanation encompassed distinct portions of the music or distinct sections within the ensemble. If the conductor restarted the ensemble and stopped again, we tallied each variable anew, even if the same matters were repeated.
Also of interest to us were variables related to certain instructional procedures evinced by the participants in the structure and content of their rehearsals, such as the number of additional performance trials (beyond the first) on a specified performance task and the number of performance episodes that began with either a directive or with no instruction. We also counted the number of full ensemble performance episodes during the rehearsals and the number of performance episodes involving only a portion of the ensemble. We calculated the duration of each of these episodes plus the percentage of total time spent in ensemble performance. These measurements are listed in Table S2 (online at http://jrme.sagepub.com/supplemental).
Two authors (who did not participate in our earlier study and who were unaware of the participants’ experimental assignments) independently observed 24 (25%) of the rehearsal recordings, selected at random, and tallied the occurrence of teacher verbalizations per our operational definitions. Durations of performance episodes in rehearsal were also calculated. Reliability for these observations ranged from .70 to perfect agreement, with an average per-participant reliability of .81. These authors then viewed the remaining 72 recordings (36 for each author).
Results
For each teacher, verbalization frequencies from the three rehearsals were totaled and considered together as a single composite rehearsal (as in Goolsby, 1997; Montemayor and Moss, 2009). Duration of composite rehearsals, excluding the “run-through” performance portions, averaged 13.3 minutes. Since rehearsal lengths varied slightly among participants, the frequencies were divided by the duration of each composite rehearsal in order to provide time-standardized measurements of these verbal behaviors. Results are reported in Table S1 (available online at http://jrme.sagepub.com/supplemental). In both conditions, Dynamics was the most frequently mentioned musical category; this accounted for over 28% of all music-related comments in the internal condition and nearly 41% of comments in the external condition. In the internal condition, positive, unspecific feedback was the most frequently occurring teaching variable, followed by positive, specific/contingent feedback; these were also the most frequent teaching variables in the external condition, although in reverse order.
Because several of these measurements exhibited nonnormal distributions, we used nonparametric independent-samples Mann-Whitney U tests in our examination of differences between experimental groups. Significant differences in the rates of these verbal behaviors were found between internal and external conditions for 1 of the 16 musical variables, with balance/blend being more frequently mentioned in the internal condition (U = 71, p = .030, r = .383). Rates for one of the eight teaching variables were also significantly different between conditions, with positive/specific feedback occurring more frequently among participants in the external group (U = 52, p = .004, r = –.507).
In order to better understand general patterns of teachers’ musical concerns in rehearsal, we assigned the musical variables to one of three possible primary factors (cf. Bergee, 1995), namely Accuracy (inclusive of breath mark, entrances/confidence, notes, rhythm/subdivision, and tempo), Expression (articulations/note length, dynamics, energy, part predominance, phrasing, style/expression), and Tone Quality (airstream/posture, balance/blend, intonation, tone, tuning). We calculated the proportion of verbalizations related to each factor as a percentage of total comments (internal, n = 321; external, n = 295). Comments related to Expression were most common in both conditions, but this performance factor appeared 10% less frequently in the internal condition (53%) than in the external condition (63%). There were 8% more comments regarding Tone Quality in the internal condition (23%) than in the external condition (15%), and the proportion of comments related to Accuracy was equivalent between internal and external conditions (23% and 22%, respectively). Overall, the distribution of teachers’ verbalizations among Accuracy, Expression, and Tone Quality were significantly different between experimental groups, χ2(2, N = 616) = 8.58, p = .014, Cramér’s V = .118.
Results of our examination of variables related to instructional procedures and the structure of performance episodes in rehearsal are reported in Table S2 (available online at http://jrme.sagepub.com/supplemental). Teacher directives (i.e., performance episodes that began with instructions to the ensemble regarding what to do differently) occurred at the rate of about one per two minutes in both conditions. Performance episodes that began without such instruction occurred more frequently in the external condition (U = 72, p = .035, r = −.374). Performance episodes also occurred with significantly greater frequency among participants in the external condition (about one episode per minute, compared to slightly slower than that in the internal group; U = 72.5, p = .036, r = −.370). These episodes also were of a significantly briefer duration (about 34 seconds on average, compared to about 43 seconds in the internal group; U = 67, p = .022, r = .406). However, the overall proportion of rehearsal time spent in ensemble performance was similar between groups (internal, 48%; external, 45%).
To assess conducting quality, we presented the soundless video recordings in different random orders to five experienced instrumental musicians who were graduate students in music education. Evaluators were unaware of participants’ experimental assignments and rehearsal orders (i.e., whether a given rehearsal recording was the first, second, or third rehearsal for that participant). Using 6-point Likert-type scales, they evaluated two characteristics of participants’ conducting, namely, clarity of gesture (which we defined for evaluators as being inclusive of elements such as clear pulse, conventional beat patterns, apparent comfort with conducting apparatus, effective use of left hand, and consistent beat style; evaluated from 1 = unclear or confusing to 6 = very clear) and expression (including gestural style and variety, showing shapes of phrases, and facial expression; from 1 = entirely unexpressive to 6 = highly expressive). Reliability, as determined by the average measure intraclass correlation coefficient, was .79 for clarity and .86 for expression. For each participant on each of their three rehearsals, we calculated the mean of the five evaluators’ scores to yield a single measurement for each of these characteristics. Although the scores for these measurements were significantly correlated with one another (r = .727, p < .001), we chose to analyze them separately because we saw them as conceivably distinct entities; a conductor could, for example, exhibit strict and clear gestures that were otherwise unexpressive or exhibit enthusiastic conducting while lacking gestural clarity.
Results of both conducting evaluations show slightly rising scores for the external group over the course of the three rehearsals and declining scores for the internal group (see Figures S1 and S2, both available online at http://jrme.sagepub.com/supplemental). For clarity, mean evaluation scores ranged from 3.68, SD = .69 (internal group, third rehearsal), to 4.10, SD = .70 (external group, third rehearsal). Expression scores were lower, ranging from 2.93, SD = .79 (internal group, third rehearsal), to 3.21, SD = .87 (external group, third rehearsal). A three-way repeated measures analysis of variance procedure for the clarity measurement, with condition (i.e., internal or external) and institution as between-subjects factors and time (i.e., rehearsal order) as the within-subjects factor, revealed no significant main effects or interactions for any of these variables. However, a test of within-subjects contrasts revealed a significant linear trend according to the time by condition interaction, F(1, 28) = 5.28, p = .029, ηp2 = .159. A separate three-way repeated measures ANOVA for the expression scores revealed a significant time by condition interaction, F(2, 56) = 3.85, p = .027, ηp2 = .121, and again, a significant linear trend, F(1, 28) = 5.70, p = .024, ηp2 = .169.
Discussion
In this study, 32 undergraduate music education majors prepared for leading a series of three band rehearsals either through intensive “internal” preparation (including activities such as listening to model recordings, singing lines of the music, and marking the score) or through an “external” process (watching rehearsal videos of expert teachers and later of their own rehearsals, setting goals for their own teaching, considering specific behaviors known to be associated with effective teaching). Results of our analysis of their rehearsals demonstrated that compared to the internal group, the external group exhibited certain behaviors associated with more effective teaching, including higher rates of positive/specific feedback, and more frequent and briefer performance segments. However, the external group more often asked for the ensemble to start without providing a directive. The internal group mentioned ensemble balance in their rehearsals more frequently than did the external group, and a higher proportion of their overall verbalizations were related to ensemble Tone Quality. Independent evaluators’ ratings of participants’ conducting revealed a significant time by condition interaction for the expression scores, with scores for the external group increasing over time and scores for the internal group decreasing.
Our finding of some significant differences between groups provides a meaningful complement to our previous examination of data generated from these same rehearsals (Silvey & Montemayor, 2014). Conceptually, our results are somewhat congruent with findings from the motor behavior learning literature (Wulf, 2013), with external focus conditions yielding the best behavioral results. That said, our adaptation of internal and external terminology reveals somewhat of a paradox: What we have identified as an “external” focus had to do with the behaviors of rehearsing and conducting—opposite, for example, of the keyboard-playing participants in research by Duke, Cash, and Allen (2011), where those who focused on the motion of their fingers (an “internal” focus) underperformed those who focused on the sound they created. Granted, the natures of the respective tasks in these disparate studies are entirely different from one another in nature and complexity. To the extent that we can make a valid comparison, our findings suggest that creating an “aural image” of a musical score through private study is in fact not analogous to attentional focus on the results of musical instruction. Ensemble rehearsing, and indeed all teaching, is fundamentally interpersonal in nature, notwithstanding the necessity of personal preparation for instruction. Accordingly, the implications of this finding suggest that for music teacher education programs, instruction in score study ought to be subordinated, or at least contextualized, within a broader array of leadership skills that include musical readiness and the social capacity to effect change in students.
We would ordinarily be reluctant to withhold score study opportunities for external group participants and to withhold video-recording reviewing opportunities for the internal group for any longer than was necessary for research purposes. Indeed, participants in the external group may yet have generated certain insights about “internal” musical issues through the process of reviewing video recordings, notwithstanding the efforts to direct their attention to distinctly teacher-focused behaviors. The converse may also be true for the internal group. Still, consistent with previous research cited in this study (see also Price & Byo, 2002), having separate approaches to rehearsal preparation in this experiment yielded some behavioral differences between groups that reflected their respective advantages. The ensemble participants in our previous study (Silvey & Montemayor, 2014) didn’t discern these differences, at least in terms of their numerical ratings of the conductors’ rehearsal effectiveness or score knowledge. In the current investigation, we found that external group participants paced their rehearsals more quickly, offering briefer and more frequent performance opportunities, and they provided positive, specific feedback at higher rates than did those in the internal group. These reflect elements of preferred temporal and behavioral profiles for rehearsals (Duke et al., 1998; Yarbrough & Madsen, 1998) and elements indicative of expert teaching (Goolsby, 1999). One relative disadvantage of an exclusively “external” approach was also seen, with this group’s higher frequency of “no instruction” rehearsal segments (i.e., segments that weren’t preceded by a teacher directive), a behavior seen more frequently among novice teachers (Goolsby, 1999). This may reflect an instructional urge to maintain a brisk rehearsal pace, perhaps inspired by self-reviews of the video recordings of their teaching. Music teacher educators should be mindful of this phenomenon when assigning similar self-evaluation tasks to preservice teachers following instructional practice sessions.
We found fewer differences that seemed to favor the internal group. This may suggest the relative difficulty new teachers might have in translating “internal” ideas into behaviors that are evinced in their rehearsal instruction. However, across all of the teacher verbalization categories we measured, these significant findings were the only ones that addressed distinctly musical concerns. Internal group participants identified ensemble balance and blend as a specific element of interest more frequently than did those in the external group, and the distribution of all of their performance-related verbalizations reflected a greater concern for overall ensemble tone quality—a characteristic that is also reflective of expert teaching (Goolsby, 1997, 1999). This finding may indicate the advantage these participants had gained, by virtue of their score study, in being able to craft a more distinct and compelling “aural image” of the particular score they were assigned to conduct. Together, these sets of findings show how different attentional focus during rehearsal preparation is consequential for novices, with near exclusive attention on either “internal” or “external” matters yielding both benefits and (perhaps more notably) deficits as compared to one another. It would seem incumbent upon music teacher educators to recognize the extent to which aspects of needed improvement in preservice teachers’ rehearsal behaviors might be reflective of misplaced attention in their preparation for instruction. Such needs would likely vary a great deal among individuals. Our profession will benefit from future research that illuminates novices’ thinking in these regards and that explores the processes by which early-career teachers develop the capacity to simultaneously attend to an array of musical and instructional concerns (Borko & Livingston, 1989; Marcum & Duke, 2015).
The proportion of teacher verbalizations regarding the primary factor of Accuracy were nearly the same between conditions. In an earlier study using a similar methodology, Montemayor and Moss (2009) found a slightly greater concern for Accuracy among participants when they had prepared for rehearsals using recorded models compared to when they had not used them. Participants in that study had a similar level of experience as did those in the current investigation; however, the ensembles they rehearsed were younger and exhibited more heterogeneous levels of musical skills. Future researchers could investigate how model-supported score study and other “internal” preparation procedures function differently for teachers when working with groups of different skill levels.
Through our independent evaluations of teachers’ conducting (i.e., evaluations provided by observers who were not themselves part of the ensemble), we saw a significant time by condition interaction on the Expression measurement, with scores for the external group increasing over time. These patterns were not seen in the evaluations of conductor score knowledge or of rehearsal effectiveness as provided by ensemble members in our previous investigation (Silvey & Montemayor, 2014). This discrepancy—that is, the seeming inability of the ensemble members to discern condition-specific differences seen by independent observers—may suggest the highly social context in which rehearsal interactions take place (cf. Duke & Henninger, 2002, p. 84). Alternatively, these differences may simply reflect the fact that the ensemble members were providing more global evaluations of the conductors’ overall rehearsal work, whereas we instructed our independent evaluators to focus specifically on elements of conducting quality. Given the trend of both the expression measurement and the clarity measurement as seen in the graphs of these data (see Figures S1 and S2 at http://jrme.sagepub.com/supplemental), we speculate that the participants’ video self-reviews, which participants completed prior to their second and third rehearsals, was the particular preparation activity that effected this improvement. By contrast, the scores of the internal group participants declined over time. Because “expression” was also favored in the external group participants’ proportion of rehearsal comments, we wonder if attention to gestural quality affects teacher verbalizations, or vice versa, or both.
Several precautions are important to note. As in our previous study (Silvey & Montemayor, 2014), we see that many aspects of the rehearsals were very similar between conditions. We found no significant differences on 28 of the 34 performance, teaching, and instructional procedures variables we measured in the current study. It is possible that the musical needs evident to the participating conductors and the participants’ choices in responding to them (and their extant skills in being able to do so) precluded any behavioral differences that might have been brought about by the experimental treatments. Although the instructional activities during these treatments were quite structured and distinct from one another, participants were afforded a large measure of freedom in particular details of their preparation (e.g., how to characterize phrases of a score under study for participants in the internal group or what particular behavioral aspects external group participants ought to choose as goals for improvement). As such, it is likely that while the treatments may have been rigorous, they may have been insufficiently prescriptive for these preservice teachers at this stage in their professional and musical development. Also, we measured participants’ behaviors over only three very brief teaching episodes, which we considered together as a single, composite rehearsal for each participant. It is possible that treatment-related differences could have been evident over a longer period or over a greater number of rehearsals. Our relatively small samples and the limitations of our experimental design do not permit us to discount entirely the possibility that the differences we did find may have been preexisting in the groups themselves (notwithstanding random assignment) rather than a function of treatment. Still, the significant linear trends seen in the conducting evaluation data may suggest otherwise.
Results from this study provide empirical support for a curricular approach to music teacher education that those in the professorate may have long known intuitively, namely, that both intensive score-based preparation and deliberate behavioral analysis are necessary in the development of strong rehearsal skills; neither is sufficient by itself. Moreover, the benefits of each approach can be measured by systematic observation but may not be discernible by general evaluations of teacher performance. Continued research regarding how particular rehearsal preparation activities relate to teacher behavior and how they relate to other variables as well—such as the skill level of the teacher and the short- and long-term changes in music performance brought about by the rehearsal process—could yield insights that prove important to conductor and teacher preparation programs.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Notes
Author Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
