Effects of accurate and inaccurate visual feedback on the tuning accuracy of high school and college trombonists

Abstract

We examined how visual feedback from electronic tuners affected trombonists’ pitch performance and tuning confidence. High school (n = 29) and college trombonists (n = 30) were tasked to play in tune with a stimulus tone (G3) recorded by a professional trombonist presented through headphones. Following each of three attempts, participants rated their confidence level that their performance was in tune. A different tuner was provided during each tuning attempt and set to one of three conditions: in-tune (A = 440Hz); flat (A = 437Hz); and sharp (A = 443Hz). These tuner conditions displayed either accurate (A = 440Hz) or inaccurate (A = 437Hz; A = 443Hz) visual feedback. Results indicated significant main effects due to tuner condition and experience level. Participants’ tuning accuracy with the sharp tuner was significantly less precise in comparison to flat and in-tune tuners. Collegiate participants performed with greater precision than high school participants in all tuner conditions. Weak relationships were observed between participants’ tuning performance and confidence ratings.

Keywords

Tuning pitch matching visual feedback perceptual feedback electronic tuner

Introduction

It is perhaps incontrovertible that the ability to perform with impeccable pitch accuracy is a goal of many performing musicians and educators. As such, the pedagogy by which this ability is developed, shaped, and achieved is relevant to all musicians, most especially to those charged with the education of developing musicians (Garofalo, 1996). Morrison and Fyk identified three component skills necessary to perform with accurate pitch: (1) pitch discrimination, a perceptual skill; (2) pitch matching, a performance skill; and (3) intonation, described as “… manipulation of pitches and intervals within a real musical context” (Morrison & Fyk, 2002, p. 184). This listing suggests that performing with appropriate pitch accuracy in varying contexts involves a conglomeration and honing of several distinct skills.

Surprisingly, a weak relationship between pitch perception and pitch performance has been a consistent theme in previous research literature (Ballard, 2011; Byo, Schlegel, & Clark, 2011; Morrison, 2000; Yarbrough, Karrick, & Morrison, 1995; Yarbrough, Morrison, & Karrick, 1997). In addition, a tendency to perform and prefer sharp pitches rather than flat pitches has been found in a number of studies (Byo & Schlegel, 2016; Geringer, 1976, 1978; Morrison, 2000; Yarbrough et al., 1995). Bregman (1994) has suggested that musicians tend to perform in a sharp manner so their individual timbre has a distinctive presence in a complex mix of other timbres. It should be noted that the tendency to perform sharp pitches has not been found unanimously, however. Ely (1992), for example, found college woodwind players performed more flat than sharp when the stimulus tone was a differing woodwind timbre.

Pitch perception might vary across instruments (Geringer, MacLeod, & Sasanfar, 2015), and interactions between timbre and pitch could provide insights into perceptual differences among instruments and participants’ experience levels (Geringer, Madsen, & Dunnigan, 2001; Krumhansl & Iverson, 1992). Varying effects due to temperament systems could also be indicative of timbre changes that result from systematic pitch compromises (Karrick, 1998; Mason, 1960).

Musicians’ tuning/aural discrimination skills likely improve with experience (Byo & Schlegel, 2016; Loosen, 1995; Micheyl, Delhommeau, Perrot, & Oxenham, 2006; Morrison, 2000; Yarbrough et al., 1995, 1997), which could be due to decreasing demands on attention resulting from increased skill acquisition (Morrison & Fyk, 2002). Pedagogues and researchers have examined the effectiveness of various listening strategies such as directing attention to acoustical beats (Miles, 1972), resultant tones (Moody, 1992), and individual partials and lower octaves (Byo et al., 2011; Byo & Schlegel, 2016; McBeth, 1972). Certain forms of technology (e.g., chromatic tuners, smartphone applications, and computer software such as SmartMusic) have also been recommended because they provide useful feedback regarding musicians’ tuning accuracy (Feldman, Contzius, & Lutch, 2016; Hopkins, 2014; Strickland, 2013). Typically, these technological tools software provide feedback to the performer in real time, immediately following the performance.

Duke characterized feedback as “ . . . any stimulus occuring coincident with or subsequent to a given behavior that the learner associates with the behavior” (Duke, 2012, p. 122). If it is incumbent upon learners to associate feedback with the antecedent or concurrent behavior, it seems that feedback competes for their attention and that some feedback could be ignored or not at all perceived. Feedback can be experienced via a variety of senses through aural and visual modalities. Performers may also perceive feedback by means of touch (haptics) or proprioception as they become aware of physical feelings in muscles and joints. Tan, Pfordresher, and Harré referred to this self-perception as “perceptual feedback” (Tan, Pfordresher, & Harré, 2010, p. 218) in which performers are aware of and perceive the results of their performance. Perceptual feedback might well be framed within the context of just-noticeable difference (JND), or the magnitude of change “that is just noticeable to the listener” (Backus, 1969, p. 85). As it relates to typical aural acuity for pitch accuracy, the JND between two pitches is likely around 6 cents for pure tones (Campbell & Greated, 1987) and around 5 cents for complex tones (Backus, 1969).

The visual information provided by a chromatic tuner functions as a form of feedback. Akin to feedback, electronic tuners engage visual and auditory modalities and are a tool used commonly in wind rehearsals to improve tuning performance (Droe, Chenoweth, & Galyen, 2011; Scherber, 2014; Silvey, 2013). When a musician performs a tone, a display on the tuner provides visual information regarding the pitch accuracy of the performed tone. Eldridge, Saltzman, and Lahav (2010) found visual feedback combined with auditory feedback (audiovisual feedback) to be more effective than auditory feedback alone. As it relates to development of pitch accuracy, Welch, Howard, and Rush (1989) found that the use of real-time visual feedback resulted in participants’ (seven-year-old children) improvement in vocal pitch-matching ability. Other findings (Banton, 1995; Platt & Racine, 1985) have suggested that experience may influence the extent to which musicians rely on visual feedback.

Some pedagogues have expressed dissent over the usage of tuners with school musicians, in part due to adherence to visual feedback above aural feedback (Gorder, 1991; Griswold, 1988). Feldman et al. (2016) explained that tuner usage risks implicitly teaching students that playing accurate pitch is a visual phenomenon, rather than an aural process that underpins nearly all ensemble performance contexts. Because tuner use is a popular choice of high school band directors (Scherber, 2014; Silvey, 2013), the outcomes of tuner use are critical to examine.

While many advances have been made in the temperament settings on modern chromatic tuners, it is nearly impossible for a tuner to give visual feedback that takes into account the myriad of possible harmonic and melodic contexts that can influence contextually-accurate pitch performance. Likely, this forces musicians to ignore visual feedback and rely solely on perceptual feedback when making judgements and musical decisions regarding pitch. Perhaps a byproduct of musical skill acquisition is an increasing confidence in one’s ability to perceive errors with less reliance on external visual feedback (Banton, 1995; Platt & Racine, 1985). In particular musical scenarios, the visual feedback from a tuner may naturally align with the musical task at hand, most especially when the task is to play in unison because the 1:1 relationship requires no contemplation of context or temperament. However, when the musical context dictates just intonation, a tone may be raised or lowered as a result of its relationship to the modal context (e.g., leading tone versus tonic) or as a function of its harmonic relationship (e.g., root versus seventh in a dominant chord). Therefore, depending on the musical context, a chromatic tuner may actually provide inaccurate visual feedback (i.e., feedback that contradicts auditory perceptual feedback).

We designed the present study to investigate how, if at all, high school and college instrumentalists’ pitch performance and tuning confidence are affected by the presence of visual feedback (accurate and inaccurate) displayed by an electronic tuner. Thus, the purpose of this study was to examine the effects of tuner condition (flat (A = 437Hz), in-tune (A = 440Hz), or sharp (A = 443Hz)) and experience level (high school or college) on instrumentalists’ tuning performance and confidence of tuning accuracy. The following research questions guided the study: (1) What are the effects of tuner condition and experience level on instrumentalists’ tuning accuracy as measured in absolute cent deviation? (2) Which performed tuning results (i.e., flat, in-tune, or sharp performance) occur as a function of tuner condition? (3) What are the relationships between instrumentalists’ tuning performance and confidence of tuning accuracy?

Method

Instrument selection and stimuli creation

For this study, we selected the trombone because of its capacity to play well above and below a target pitch due to the moveable slide, which enables fine pitch adjustments without extreme manipulations of physical characteristics such as embouchure. After choosing the trombone, we considered numerous possible stimulus pitches to which the participants would tune. We chose the pitch G3 (fourth-space, bass clef) as the stimulus tone for the primary task and A3 (top line, bass clef) for the practice task because these pitches were in a comfortable playing range yet allowed room for error above/below the pitch based on slide placement.

While viewing a chromatic tuner set to A = 440Hz, a university faculty member with a completed doctoral degree in trombone performance played numerous sustained repetitions of both pitches (G3 and A3) for stimuli creation. Stimuli were recorded using a Zoom H2n Handy Recorder and exported as .wav files (44,100 Hz sampling rate, 16-bit stereo). We cropped the recorded tones and evaluated all iterations of each tone based on a mean frequency analysis using Praat software (Boersma & Weenink, 2014). We analyzed only the middle three seconds of each tone due to the expected pitch fluctuations that occur at the attack and release – a practice used in prior research (Byo et al., 2011; Byo & Schlegel, 2016).

After evaluating all performed iterations, we chose the performance of each tone that we deemed most even in quality that closely matched the equal temperament standard frequencies and fell within established 5-cent JND thresholds for complex tones (Backus, 1969). The selected stimuli had mean frequencies of 195.9 Hz for G3 (0.86 cents below the equal temperament standard) and 219.5 Hz for A3 (3.93 cents below the equal temperament standard). We then prepared two stimulus tracks, one using the G3 pitch and the other using A3. Each stimulus tone was approximately 5 seconds in length. Both stimulus tracks included ten repetitions of the stimulus tone, which were separated by 2-second silence intervals.

Participants and procedures

We conducted a power analysis using G*Power software version 3.1.9.2 (Faul, Erdfelder, Lang, & Buchner, 2007) to determine a sufficient sample size for repeated measures analysis of variance (ANOVA) with a within–between interaction¹, the results of which suggested a minimum sample size of 58. Participants (n = 59) were collegiate trombonists (n = 30) sampled from two National Association of Schools of Music-accredited universities in the United States, one in the Southeast and one in the Northwest, and high school trombonists (n = 29) sampled from five high schools in the United States, three in the Southeast and two in the Northwest. The mean age of the high school group was 15.83 years (SD = 1.01), and mean age of the collegiate group was 21.54 years (SD = 2.99). High school participants reported a mean of 5.21 years of playing experience (SD = 1.40), and collegiate participants reported a mean of 10.87 years of playing experience (SD = 2.94). All collegiate trombonists were music majors. Participants met with one of us individually to complete the institutional review board-approved research procedures.

The primary research question in this study concerned the potential effects of accurate and inaccurate visual feedback as displayed by an electronic tuner on participants’ tuning accuracy. To create varying tuner conditions, we calibrated three Korg CA1 chromatic tuners to different frequency (Hz) levels: A = 437Hz (“flat” tuner), a condition that displayed inaccurate feedback; A = 440Hz (“in-tune” tuner), a condition that displayed accurate feedback; and A = 443Hz (“sharp” tuner), a condition that displayed inaccurate feedback. If playing in tune, the flat tuner would have indicated to the participants that their performance was sharp, the in-tune tuner would have indicated that their performance was in tune, and the sharp tuner would have indicated that their performance was flat. The 3-Hz adjustments with these notes in this octave resulted in an 11.8-cent deviation above and below the equal-tempered standard for the two inaccurate tuner conditions. This manipulation, specifically in the “flat” and “sharp” tuner conditions, created a situation in which participants would likely encounter incongruent visual and auditory perceptual feedback.

Participants heard the trombone stimulus through AUVIO 33-279 Concert Class stereo headphones. The wearing of the headphones enabled us to isolate, capture, and eventually analyze only the participants’ tuning attempts – a practice used in previous research (Byo et al., 2011; Byo & Schlegel, 2016; Yarbrough et al., 1995, 1997). Participants played into a Shure SM 57 microphone. Both the headphones and the microphone were connected to a Behringer XENYX802 mixer so participants would be able to hear themselves and the stimulus pitch simultaneously. Participants’ performances were recorded with a Zoom H2n Handy Recorder.

Before collecting any data, participants were acquainted with all procedures through a practice experience. All instructions and procedures in the practice experience were identical to those used in the primary tasks, with the only difference being the musical stimuli that were utilized in this experience (the pitch A3 for the practice task; G3 for the primary tasks). The practice experience not only oriented participants to the procedures but also enabled us to make any adjustments in the setup of equipment and volume levels of the stimulus tone.

Prior to tuning, all participants were allotted time to warm-up their instrument. After the warm-up, we read the procedural instructions to the participants. These instructions were printed on a distinctly-colored performance instruction sheet, and the stimulus pitch was notated as a whole note on a bass clef staff below the instructions. After hearing the instructions, participants sustained the stimulus pitch (A3 in practice task; G3 in primary tasks) several times until they believed that they were in tune with the stimulus sounding in their headphones. These are the procedural instructions we read to participants:

Please perform “in tune” with the trombone stimulus sounding in the headphones. A tuner is provided for your reference and as an aid to your performance. You can play the (A3 practice task; G3 primary tasks) several times until you believe you are in tune with the trombone stimulus in your headphones. Once you believe you are in tune with the stimulus in your headphones, articulate that (A3 practice task; G3 primary tasks) again and sustain it for about five seconds. Please do articulate each (A3 practice task; G3 primary tasks) you play.

After completing the practice experience, participants tuned three times, once in each tuner condition. Three presentation orders were used, and each tuner condition was presented first, middle, and last across the three orders to ensure contrasting presentations; orders were randomly assigned to participants. During each tuning attempt, a different Korg CA-1 tuner was placed on the stand beside the performance instruction sheet. The Hertz (Hz) setting on each tuner was covered with a sticker that matched the color of the corresponding performance instruction sheet. The colored sticker prevented participants from viewing the Hz settings on the tuners and functioned as a procedural and organization element. In the practice experience, the in-tune tuner was provided to avoid priming the participants to play out of tune. It is important to note that we read the procedural instructions to participants before each of the tuning tasks; this was the extent to which we directed participants’ attention to the tuner. We tested all these procedures during a pilot study and held debriefing sessions with each pilot participant. None of the pilot participants expressed any concerns or questions regarding the toggling procedure used with color-coded instruction sheets and tuners. Furthermore, we observed no unusual responses or questions/concerns from participants due to these procedural elements in the main study.

Following each tuning task, we asked participants, “On a scale of 1 (not at all confident) to 6 (very confident), how confident are you that you played in tune with the stimulus you heard in your headphones?” We chose an even-numbered rating scale to eliminate the possibility of a neutral response. After each tuning task and confidence-rating question, participants performed a scalar passage (E-flat major in practice experience/D-flat major in primary tasks) in order to limit tonal memory of the stimulus. Because the scalar passage did not contain the stimulus tone and its tonic was distanced from the stimulus by a tritone, we determined it to be an adequate distraction task to limit tonal memory of the stimulus.

The dependent measures in this investigation were: (1) participants’ tuning accuracy as measured in absolute cent deviation; and (2) participants’ confidence in their tuning accuracy as measured through self-report described above. We measured cent deviation by analyzing the last G3 participants played before indicating that they were in tune. These measurements were completed for each iteration of the tuner conditions, resulting in three tones for analysis per participant.

To measure mean frequency of each tone, we used Praat software (Boersma & Weenink, 2014), which analyzed the pitch (in Hz) of the sound file at a sampling rate of 100 times per second. The software averaged these values across the entire sound file and provided the mean frequency. This value in Hz was compared to 195.9Hz, the mean frequency of the trombone stimulus tone that participants were instructed to match. The cent deviation value was determined using a “cent value-determination of an interval” calculator (Sengpiel, 2014) and was interpreted in absolute value for the primary analysis.

Results

Participants’ cent deviation scores across three orders were tested for possible order effects. Results indicated no significant effects due to order (p = 0.56), and no significant interactions among order and other variables were found (order by tuner condition, p = 0.59; order by experience level, p = 0.77; order by tuner condition by experience level, p = 0.82). Therefore, all data were considered as one data set, and order was excluded in future analyses.

The first research question was focused on the effects of tuner condition (flat (A = 437Hz), in-tune (A = 440Hz), or sharp (A = 443Hz)) and experience level (high school or college) on participants’ tuning accuracy. To address this question, we conducted a mixed, factorial ANOVA using absolute cent deviation values across the three tuner conditions as the within-subjects factor and experience level as the between-subjects factor. Results indicated a significant main effect due to tuner condition, F(2, 114) = 9.97, p < 0.001, η_p² = 0.15, and the effect size was modest (Cohen, 1988). A Bonferroni adjustment for multiple comparisons indicated that participants’ absolute cent deviation was significantly greater when provided the sharp tuner (M = 8.51 cents, SD = 6.51) in comparison to their tuning accuracy when provided the flat tuner (M = 5.96 cents, SD = 6.00, p = 0.002) and the in-tune tuner (M = 5.37 cents, SD = 4.88, p = 0.001). The in-tune and flat tuner condition means were not significantly different from each other (p > 0.05).

In addition, experience level also had a significant main effect on participants’ tuning accuracy, F(1, 57) = 6.10, p = 0.017, η_p² = 0.10, and the effect size was also modest (Cohen, 1988). High school participants’ performed cent deviation (M = 8.10 cents, SD = 5.70) was significantly greater than college participants’ deviation (M = 5.16 cents, SD = 5.82), indicating that high school participants performed more out of tune than college participants overall. Although both main effects (tuner condition and experience level) were significant, the interaction between them was not, F(2, 114) = 0.57, p = 0.57. Means for both experience levels across these tuner conditions are illustrated in Figure 1.

Figure 1.

Mean absolute cent deviation across tuner conditions and experience levels. Error bars represent standard errors.

To answer the second research question regarding performed tuning results, we coded participants’ responses as sharp, flat, or in tune based on the direction to which they deviated from the mean frequency of the trombone stimulus (195.9Hz). Similar to previous research in this area (Byo et al., 2011; Byo & Schlegel, 2016), a participant’s performance was categorized as “in tune” if the cent deviation value was no more than five cents below or above the standard. A cent deviation value greater than five cents above the standard was categorized as “sharp,” and a cent deviation value greater than five cents below the standard was categorized as “flat.” This five-cent range above and below the target pitch is consistent with just noticeable difference thresholds described by Backus (1969).

These categorical data are summarized in Figure 2 for each performance in the three tuner conditions, with a total of 177 categorized performances. Across all participants (n = 59), there were 31 flat performances (17.5%), 59 sharp performances (33.3%), and 87 in-tune performances (49.2%). High school and college participants had a similar number of performances that were flat (N_HS = 14; N_COL = 17) but were less similar in the number of performances that were sharp (N_HS = 37; N_COL = 22) and in-tune (N_HS = 33; N_COL = 54).

Figure 2.

Number of performed tuning results by experience and tuner condition.

Also visible in Figure 2 are the different performance responses as a function of tuner condition. Both the flat and sharp tuner condition provided inaccurate visual feedback. In the flat tuner condition, a flat performance could indicate that a participant was influenced by the inaccurate visual feedback from the tuner. In the sharp tuner condition, a sharp performance could indicate a participant was influenced by the inaccurate visual feedback from the tuner. In the in-tune tuner condition, the visual feedback was accurate; therefore, any out-of-tune performance could suggest a participant either did not perceive a pitch-matching problem, did not look at the visual information, or both. In general, these descriptive data suggest that high school trombonists were most influenced by the visual feedback and that all participants performed sharp more often than flat, except in the flat tuner (A = 437Hz) condition.

Following each of the three tuning attempts, participants indicated how confident they were that their performance was in tune with the stimulus they heard in the headphones. To answer the third research question, we calculated the Spearman correlation (due to ordinal nature of the rating scale) between these confidence ratings and their performed absolute cent deviation values in each tuner condition. All correlation values, which were relatively small in magnitude, are displayed in Table 1. There were no significant relationships between high school participants’ confidence ratings and their tuning performance across the three tuner conditions. In contrast, collegiate trombonists’ relationship between their confidence ratings and cent deviation values was significant in the flat tuner condition (r_{s =} -0.44, p = 0.02). This instance was the only significant correlation across the conditions for both groups.

Table 1.

Spearman correlations illustrating relationships between confidence ratings and performed cent deviation across tuner conditions.

	High school (n = 29)		College (n = 30)
	r _s	p	r _s	p
A = 437Hz	0.13	0.51	−0.44	0.02*
A = 440Hz	−0.15	0.46	−0.36	0.05
A = 443Hz	−0.17	0.38	0.04	0.83

Notes: Participants rated how confident they were that they played in tune with the stimulus on a scale anchored by 1 (not at all confident) and 6 (very confident); *p < 0.05.

Discussion

The primary purpose of this study was to examine the effects of tuner condition and experience level on instrumentalists’ pitch matching accuracy and tuning confidence. Both tuner condition and experience level had significant but modest main effects on trombonists’ performance, and no interaction between these factors was observed. Collegiate trombonists performed with greater accuracy than high school trombonists across all tuner conditions (see Figure 1), a result we expected because a positive linear relationship between pitch performance accuracy and experience is prominent in the prior literature (Byo & Schlegel, 2016; Loosen, 1995; Micheyl et al., 2006; Morrison, 2000; Yarbrough et al., 1995, 1997). This significant main effect due to experience, along with the absence of an interaction between experience and tuning accuracy, indicates that the influence of the tuner on participants’ tuning accuracy was similar across conditions.

Findings in the present study support two prominent themes in the literature: a sharp performance tendency (Byo & Schlegel, 2016; Geringer, 1976, 1978; Morrison, 2000; Yarbrough et al., 1995) and a weak relationship between tuning performance and perception tasks (Byo et al., 2011; Byo & Schlegel, 2016; Geringer, 1976, 1978; Morrison, 2000). Cent deviation values in the sharp tuner condition were significantly greater than the in-tune and flat tuner conditions. This finding suggests that musicians might seek to avoid flatness, but as shown in Figure 2, this conclusion does not provide a comprehensive explanation of the phenomenon because only a few trombonists – five college and three high school – performed sharp in the presence of the flat tuner. It seems that these trombonists demonstrated more than just a tendency to avoid flatness and that their performance may have been influenced by the visual feedback displayed by tuners, although they were influenced by the sharp tuner differently than the in-tune and flat tuners.

Also displayed in Figure 2, the frequency of sharp tuning results in the sharp tuner condition (A = 443Hz) is strikingly greater than the frequency of flat results in the flat tuner condition (A = 437Hz). This result is uniquely meaningful because both of these tuner conditions displayed inaccurate visual feedback of equal deviation from the equal-tempered standard. It was in the sharp tuner condition where the inaccurate feedback resulted in more instances of inaccurate tuning performances.

Correlations between participants’ confidence ratings (i.e., confidence that they played in tune with the aural stimulus) and their performed absolute cent deviation illustrated a weak relationship between actual performance and confidence of performance accuracy on this tuning task. There were no significant relationships among high school participants, and only the relationship between confidence ratings and cent deviation in the flat tuner condition was significant for college participants. It should be noted that the relationship between participants’ performance and their confidence of tuning accuracy was stronger among the collegiate participants, with the exception of their performance in the presence of the sharp tuner. The nature of the negative correlations, specifically for the flat and in-tune tuner conditions, is worthy of contemplation. This inverse relationship suggests that as cent deviation values increased, confidence ratings decreased and vice versa (i.e., more out of tune playing was related to lower confidence ratings). As presented in Table 1, not all relationships demonstrated this expected pattern; two positive correlations were observed, indicating the reverse trend. These inconsistent trends in the data further support prior research indicating a weak relationship between tuning perception and performance (Ballard, 2011; Byo et al., 2011; Morrison, 2000; Yarbrough et al., 1995). Despite the nature of the correlations, the strength of the association between tuning accuracy and confidence was quite small across participants.

The tendency to perform sharp and inaccurately perceive one’s own performance indicates a need to hone the individual skills suggested by Morrison and Fyk (2002) – pitch discrimination, pitch matching, and intonation. The tendency could be due to certain practical and physical variables (e.g., lack of experience or physical tension). Bregman suggested another possibility – that sharp performance might be a natural consequence of participants’ tendency to want/need to distinctly hear their instrumental timbre:

The simplest way to keep [sounds] distinct is to have them sing or play notes that are different from the other parts. This is effective because the ear is trying to find a set of different fundamental frequencies to which the detected partials can be allocated. This search by the ear for related harmonics explains why instrumentalists, in order to avoid having their sounds blend with the others, will sometimes play slightly sharp. (Bregman, 1994, p. 490)

How can visual feedback function as a scaffold to encourage student growth related to certain desirable skills such as perceiving in-tuneness and performing in tune? Maybe the mere presence of a tuner generates an awareness, a cue that encourages the ear to “pay attention” to the nuance and details of pitch performance. Morrison and Fyk (2002) expounded on the notion of pitch accuracy as a function of experience. They suggested that less-skilled instrumentalists have a myriad of variables to attend to, which thereby limits their attentional resources when completing such complex tasks as playing in tune.

It is worth noting that although we told participants in the instructions that a tuner was available, it was their choice whether or not to view the tuner and the feedback it displayed while performing. We decided to not overly direct participants’ attention to the tuner for several reasons. First, we were mindful of the fact that if we attempted to overly direct their attention to the tuner, then we would risk exposing the general hypothesis and purpose of the investigation, which would confound the results. Second, because humans are constantly presented with feedback of various types and modalities (Duke, 2012), it is possible that some feedback could go unnoticed or even be ignored. Third, musicians use electronic tuners with the assumption that the feedback displayed is accurate and appropriate to the musical context. As such, the goal of this inquiry was to examine how available visual feedback from the tuner would affect participants’ tuning accuracy, especially when this visual feedback could be in conflict with auditory perceptual feedback.

Based on these results, we offer several recommendations to wind music educators. First, we recommend that teachers regularly provide sound models of in-tune and out-of-tune performances. Because in-tune and out-of-tune performances are sound concepts (Byo & Schlegel, 2016), students will need regular exposure to these concepts in order to create accurate memories of these sounds. Until musicians know which feedback in the sound environment to attend to, they may solely depend on what they can see from the tuner or what they are told from an instructor. In this study, participants’ performance, especially in the sharp tuner condition, suggested a lack of awareness regarding the physical and auditory byproducts of performing out of tune.

Second, we encourage teachers to use chromatic tuners for the purpose of adjusting instrument length at the beginning of rehearsal (i.e., during mass ensemble tuning procedures when instrumentalists tune their instrument by adjusting instrument length) or for periodic assessment of instrument length. Use of chromatic tuners may also be beneficial for students to identify the pitch tendencies of their instrument, although there are several variables that affect pitch accuracy (e.g., embouchure and temperature) that may interact with instrument length. We also recommend that the visual feedback from chromatic tuners be used in conjunction with audible pitch models, where the pitch the student is trying to match is being modeled by the teacher, another student, or even a recording. Pairing visual and auditory feedback creates audiovisual feedback, which has been found to be more efficacious than visual feedback alone for long-term skill development and pitch memory (Eldridge et al., 2010).

Third, we recommend that teachers give students opportunities to practice pitch discrimination and pitch matching separately because these appear to be distinct skills (Morrison & Fyk, 2002), especially with developing instrumentalists. While pitch discrimination is an important skill, it may not be predictive of the ability to play in tune (Ballard, 2011; Byo et al., 2011; Morrison, 2000; Yarbrough et al., 1995, 1997), possibly a result of interactions among variables affecting pitch accuracy, such as embouchure, body/hand position, and other issues. Perhaps pitch discrimination and pitch accuracy become more strongly correlated and predictive of each other as a function of increasing skill (Loosen, 1995; Micheyl et al., 2006; Morrison, 2000; Yarbrough et al., 1995, 1997) and exposure to the sound concept of “in tune” performance (Byo & Schlegel, 2016).

We acknowledge certain limitations of the study that should be considered. Some participants indicated after completing the tasks that they typically do not use a tuner when matching pitch. We observed that some participants closed their eyes and only glanced intermittently at the tuner. We also observed the opposite as well, as some participants focused overtly on the tuners while others merely glanced between the tuner and the printed instruction sheet. Because we did not have a control (no tuner) condition in which participants could only rely on their auditory perceptual feedback, we cannot know how much participants relied on the visual feedback in this pitch-matching task. Despite this limitation, it is important to consider that the only variable altered across the three primary tuning tasks was the tuner frequency setting. Participants heard the same stimulus in their headphones across the tasks, and due to the lack of an order effect in the presentation of the tuners, the logic of causation leads us to conclude that the manipulated variable may have been one cause for significant differences in performance. In future studies, investigators should consider including a control condition in which participants tune to an aural stimulus without any visual feedback, which would allow a more inclusive investigation of the effects of visual feedback on tuning accuracy.

Additionally, the performance tasks in this study involved the analysis of single tones, rather than the analysis of a larger-scale musical idea (i.e., one in a harmonic and/or melodic context). We felt that this limited musical context was appropriate for the present study because the task of tuning to a single stimulus tone is an ecologically valid performance task – one that regularly occurs as a means of calibrating musicians’ instrument lengths (and their aural expectations) prior to rehearsals and performances (Droe et al., 2011; Fabrizio, 1994; Garofalo, 1996).

In future studies, the use of larger-scale musical ideas within a harmonic and/or melodic context would extend the findings of the present study to different music performance situations. Follow-up studies examining pitch performance of instruments that manipulate pitch using different physical means (e.g., bowed string instruments) could also provide additional insight into musicians’ tuning behaviors. Given the ubiquitous nature of intonation as a performance element in ensemble performance contexts (Fabrizio, 1994; Garofalo, 1996), future research in these areas would provide further insight into ways that school musicians, and even advanced musicians, can be made aware of pitch accuracy.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

References

Backus

(1969). The acoustical foundations of music. New York, NY: W. W. Norton & Company, Inc.

Ballard

D. L.

(2011). Relationships between college-level wind instrumentalists’ achievement in intonation perception and performance. Bulletin of the Council for Research in Music Education, 187(1), 19–32.

Banton

L. J.

(1995). The role of visual and auditory feedback during the sight-reading of music. Psychology of Music, 23(1), 3–16.

Boersma

Weenink

(2014). Praat (Version 5.3.80) [Computer software]. Retrieved from http://www.praat.org/ (accessed October 4, 2015 ).

Bregman

A. S.

(1994). Auditory scene analysis: The perceptual organization of sound. Cambridge, MA: MIT press.

Byo

J. L.

Schlegel

A. L.

(2016). Effects of stimulus octave and timbre on the tuning accuracy of advanced college instrumentalists. Journal of Research in Music Education, 64(3), 344–359.

Byo

J. L.

Schlegel

A. L.

Clark

N. A.

(2011). Effects of stimulus octave and timbre on the tuning accuracy of secondary school instrumentalists. Journal of Research in Music Education, 58(4), 316–328.

Campbell

Greated

(1987). The musician’s guide to acoustics. New York, NY: Schirmer.

Cohen

(1988). Statistical power analysis for the behavioral sciences. 2nd edition. New York, NY: Academic Press.

10.

Droe

K. L.

Chenoweth

Galyen

S. D.

(2011). An examination of before and after rehearsal pitch tendencies of string and wind musicians. String Research Journal, 2(1), 25–35.

11.

Duke

R. A.

(2012). Intelligent music teaching: Essays on the core principles of effective instruction. 9th printing. Austin, TX: Learning and Behavior Resources.

12.

Eldridge

Saltzman

Lahav

(2010). Seeing what you hear: Visual feedback improves pitch recognition. European Journal of Cognitive Psychology, 22(7), 1078–1091.

13.

Ely

M. C.

(1992). Effects of timbre on college woodwind players’ intonational performance and perception. Journal of Research in Music Education, 40(2), 158–167.

14.

Fabrizio

A. C.

(1994). A guide to the understanding and correction of intonation problems. Delray Beach, FL: Meredith Music

15.

Faul

Erdfelder

Lang

A. G.

Buchner

(2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175–191.

16.

Feldman

Contzius

Lutch

(2016). Instrumental music education: Teaching with the musical and practical in harmony. 2nd edition. New York, NY: Routledge.

17.

Garofalo

R. J.

(1996). Improving intonation in band and orchestra performance. Ft. Lauderdale, FL: Meredith Music.

18.

Geringer

J. M.

(1976). Tuning preferences in recorded orchestral music. Journal of Research in Music Education, 24(4), 169–176.

19.

Geringer

J. M.

(1978). Intonational performance and perception of ascending scales. Journal of Research in Music Education, 26(1), 32–40.

20.

Geringer

J. M.

Madsen

C. K.

Dunnigan

(2001). Trumpet tone quality versus intonation revisited: Two extensions. Bulletin of the Council for Research in Music Education, 148(1), 65–78.

21.

Geringer

J. M.

MacLeod

R. B.

Sasanfar

J. K.

(2015). In tune or out of tune: Are different instruments and voice heard differently? Journal of Research in Music Education, 63(1), 89–101.

22.

Gorder

W. D.

(1991). Intonation and sonority. Band Directors’ Guide, 5(4), 6–8.

23.

Griswold

H. E.

(1988). How to teach aural skills with electronic tuners. Music Educators Journal, 74(5), 49–51.

24.

Hopkins

M. T.

(2014). Pilot-testing of new software for measuring string players’ instrument tuning skills. Journal of Music, Technology, & Education, 7(1), 5–21.

25.

Karrick

(1998). An examination of the intonation tendencies of wind instrumentalists based on their performance of selected harmonic musical intervals. Journal of Research in Music Education, 46(1), 112–127.

26.

Krumhansl

C. L.

Iverson

(1992). Perceptual interactions between musical pitch and timbre. Journal of Experimental Psychology: Human Perception and Performance, 18(3), 739–751.

27.

Loosen

(1995). The effect of musical experience on the conception of accurate tuning. Music Perception, 12(3), 291–306.

28.

Mason

J. A.

(1960). Comparison of solo and ensemble performance with reference to Pythagorean, just, and equi-tempered intonation. Journal of Research in Music Education, 8(1), 31–38.

29.

McBeth

W. F.

(1972). Effective performance of band music. San Antonio, TX: Southern Music.

30.

Micheyl

Delhommeau

Perrot

Oxenham

A. J.

(2006). Influence of musical and psychoacoustical training on pitch discrimination. Hearing Research, 219(1–2), 36–47.

31.

Miles

(1972). Beat elimination as a means of teaching intonation to beginning instrumentalists. Journal of Research in Music Education, 20(4), 496–500.

32.

Moody

G. E.

(1992). Using resultant tones to aid intonation. NACWPI Journal, 40(4), 9–12.

33.

Morrison

S. J.

(2000). Effect of melodic context, tuning behaviors, and experience on the intonation accuracy of wind players. Journal of Research in Music Education, 48(1), 39–51.

34.

Morrison

S. J.

Fyk.

. (2002). Intonation. In McPherson

Parncutt

(Eds.), The science and psychology of music performance: Creative strategies for teaching and learning (pp. 183–198). New York, NY: Oxford University Press

35.

Platt

J. R

Racine

R. J.

(1985). Effect of frequency, timbre, experience, and feedback on musical tuning skills. Perception and Psychophysics, 38(6), 543–553.

36.

Scherber

R. V.

(2014). Pedagogical practices related to the ability to discern and correct intonation errors: An evaluation of current practices, expectations, and a model for instruction. Doctoral Dissertation. Florida State University, Tallahassee, USA. Retrieved from https://diginole.lib.fsu.edu/islandora/object/fsu:254498/datastream/PDF/view (accessed September 21, 2015 ).

37.

Sengpiel

(2014). Sengpiel audio. Retrieved from http://www.sengpielaudio.com/calculator-centsratio.htm (accessed December 11, 2015 ).

38.

Silvey

B. A.

(2013). Missouri high school band directors’ reports of tuning procedures, warm-up materials, and rehearsal time. Missouri Journal of Research in Music Education, 50(1), 20–32.

39.

Strickland

K. E.

(2013). The effect of real-time pitch tracking and correction on high school instrumentalists’ tuning accuracy. Doctoral Dissertation. Louisiana State University, Baton Rouge, USA. Retrieved from https://digitalcommons.lsu.edu/cgi/viewcontent.cgi?article=2893&context=gradschool_dissertations (accessed October 2, 2015 ).

40.

Tan

S. L.

Pfordresher

Harré

(2010). Psychology of music: From sound to significance. London, UK: Routledge and Psychology Press.

41.

Welch

G. F.

Howard

D. M.

Rush

(1989). Real-time visual feedback in the development of vocal pitch accuracy. Psychology of Music, 17(2), 146–157.

42.

Yarbrough

Karrick

Morrison

S. J.

(1995). Effect of knowledge of directional mistunings on the tuning accuracy of beginning and intermediate wind players. Journal of Research in Music Education, 43(3), 232–241.

43.

Yarbrough

Morrison

S. J.

Karrick

(1997). The effect of experience, private instruction, and knowledge of directional mistunings on the tuning performance and perception of high school wind players. Bulletin of the Council for Research in Music Education, 134(1), 31–42.