Effects of Gesture Height on Individual and Ensemble Singing: Acoustic and Perceptual Measures Lesley Maxwell Mann

Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2014 Effects of Gesture Height on Individual and Ensemble Singing: Acoustic and Perceptual Measures Lesley Maxwell Mann

Follow this and additional works at the FSU Digital Library. For more information, please contact [email protected] FLORIDA STATE UNIVERSITY

COLLEGE OF MUSIC

EFFECTS OF GESTURE HEIGHT ON INDIVIDUAL AND ENSEMBLE SINGING: ACOUSTIC

AND PERCEPTUAL MEASURES

LESLEY MAXWELL MANN

A Dissertation submitted to the College of Music in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Degree Awarded: Summer Semester, 2014 Lesley Maxwell Mann defended this dissertation on May 2, 2014. The members of the supervisory committee were:

John. M. Geringer Professor Directing Dissertation

David Okerlund University Representative

André J. Thomas Committee Member

Judy K. Bowers Committee Member

Kevin Fenton Committee Member

The Graduate School has verified and approved the above-named committee members, and certifies that the dissertation has been approved in accordance with university requirements.

“He loved her, of course, but better than that, he chose her, day after day. Choice: that was the thing.” – Sherman Alexie

To my husband, who daily chooses love and respect, and demonstrate such with his actions.

iii

ACKNOWLEDGMENTS

“What you do speaks so loudly that I cannot hear what you say.” - Ralph Waldo Emerson

This dissertation is the culmination of an effort made possible by the support of many. The completion of a doctoral degree truly takes a village, and I am forever indebted to the faculty at FSU. The culture of support at Florida State University honors engagement and encouragement, and fosters an unsurpassed community of collegiality. To my major professor, Dr. Geringer, I thank you for your guidance and wisdom in this endeavor. Your patience and thoughtfulness served as a consistently calming influence in the storm. To Dr. Thomas, what a journey! In the choral world, there are conductors who teach, and teachers who conduct, but you are gifted in the highest magnitude in both areas. I am so thankful for your ability to read what I have needed in countless situations, and for your candor and kindness in its delivery. To Dr. Bowers, I cannot imagine a better mentor for this profession. There can be no doubt of your impact on my teaching. Your legacy in my life is to seek excellence in music, teaching, performance, and service. To Dr. Fenton, I am forever grateful for the flame you rekindled in me for artistry in music making. My time in the Festival Singers of Florida reminded me of the kind of musician I can be, and I would not be here without that impetus. To Professor Okerlund, thank you for your willingness to take on the choral folk, and to have a candid conversation about voice use in all manner of settings. To Dr. Madsen, your lessons pervade all manners of my life. I am grateful, and I intend to pass them on. For their assistance in technical aspects of this project, I extend my gratitude to Eitaro Kawaguchi for his engineering of the head-mounted microphones, and to Chandler Bridges for his help with recording. A special thanks to the Gentex Corporation for their generous donation of microphone elements. To my colleagues in the graduate choral conducting program, I thank you for your friendship, inspiration, and laughter. I look forward to a long history of collaboration and conversation. To my parents and sister: I am uniquely blessed to have grown up in your love and according to your moral code. Your actions and words have demonstrated your steadfast faith and support of my endeavors. Thank you. To my children, your patience in my many absences over the last three years is a tribute to your resilient spirits. I hope that these traits accompany you for the rest of your lives. My love and thanks to you who are my great joy in life. To my husband, this journey constituted a great risk, with no guarantee of great reward. Thank you for living the dream with me.

TABLE OF CONTENTS

List of Tables ...... vii List of Figures ...... viii Abstract ...... ix 1. INTRODUCTION ...... 1 Need for the Study ...... 3 Purpose Statement ...... 4 Delimitations ...... 4 2. REVIEW OF LITERATURE ...... 5 Pedagogical Practices in Conducting ...... 5 Conductor Posture ...... 5 Hand and Arm Position ...... 6 Conducting Plane and Size of Beat ...... 6 Verbal and Nonverbal Communication in Music ...... 7 Time Use in Rehearsals ...... 8 Kinesics ...... 10 Mimicry ...... 11 Gesture ...... 12 Conducting Emblems ...... 14 Conducting Gesture in Choral Performance ...... 15 Intonation ...... 16 Tone Quality ...... 18 Vibrato ...... 21 Solo and Choral Singing Modes ...... 25 Singer Health ...... 26 Summary ...... 27 3. METHOD ...... 28 Phase One: Preparing the Stimulus Videos ...... 28 Conductor Selection ...... 28 Independent Variables: Gesture Height and Sung Vowels ...... 28 Musical Selection ...... 29 Creating the Stimuli ...... 29 Phase Two: Participant Performance ...... 31 Participants ...... 31 Phase Two Recording Venue ...... 32 Phase Two Equipment ...... 32 Phase Two Pilot Testing ...... 33 Phase Two Procedures ...... 33 Phase Two Analysis ...... 36 Phase Three: Perceptual Analysis ...... 37 Audio Recordings for Expert Panel ...... 37 Phase Three Procedures ...... 40

4. RESULTS ...... 42 Purpose Statement ...... 42 Research Question One: Singer Performance ...... 42 Singer Participant Demographics ...... 43 Performance Analysis ...... 43 Intonation ...... 43 Vibrato Rate ...... 45 Vibrato Extent ...... 46 Vibrato Duration ...... 49 Tone Quality ...... 50 Research Question Two: Singer Perceptions ...... 51 Research Question Three: Expert Listener Perceptions ...... 52 Expert Listener Reliability ...... 53 Analysis of Ensemble Performance ...... 53 Analysis of Individual Performance ...... 54 5. DISCUSSION ...... 58 Purpose Statement ...... 58 Summary of Results ...... 58 Question 1. Singer performance in response to gesture height ...... 58 Question 2. Singer perceptions of the effect of gesture ...... 58 Question 3. Expert listener perceptions ...... 59 General Discussion ...... 59 Question 1. Singer Performance in Response to Gesture Height ...... 59 Do singers change intonation? ...... 59 Do singers change vibrato performance? ...... 60 Do singers change tone quality (spectral center of gravity)? ...... 61 Question 2.: Singer Perceptions of the Effect of Gesture ...... 62 Question 3. Expert Listener Perceptions ...... 63 Recommendations for Future Research ...... 64 Conclusion ...... 65 APPENDICES ...... 66 A. HUMAN SUBJECTS COMMITTEE APPROVAL MEMORANDUM ...... 66 B. APPROVED SINGER PARTICIPANT CONSENT FORM ...... 67 C. INDIVIDUAL MICROPHONE SPECIFICATIONS ...... 68 D. SINGER PARTICIPANT QUESTIONNAIRE ...... 69 E. EXPERT LISTENER RATING SHEET ...... 70 F. PARTICIPANT VERBATIM COMMENTS (GENERAL) ...... 87 G. PARTICIPANT VERBATIM COMMENTS (SPECIFIC) ...... 93

REFERENCES ...... 98 BIOGRAPHICAL SKETCH ...... 110

LIST OF TABLES

3.1 Phase 2 Experimental Stimuli by Gesture Height and Vowel ...... 35

3.2 Phase 3 Experimental Stimuli for Individual Recordings ...... 40

3.3 Order of Expert Listener Recordings ...... 41

4.1 Means and Standard Deviations of Intonation Scores according to Gesture Height and Vowel ...... 44

4.2 Means and Standard Deviations of Vibrato Rates according to Gesture Height and Vowel ...... 46

4.3 Means and Standard Deviations of Vibrato Extent according to Gesture Height and Vowel ...... 47

4.4 Means and Standard Deviations of Vibrato Duration according to Gesture Height and Vowel ...... 49

4.5 Means and Standard Deviations of Spectral Center of Gravity according to Gesture Height and Vowel ...... 50

4.6 Frequency of Musical Elements Listed in Response to Participant Questionnaire ...... 52

4.7 Means and Standard Deviations of Expert Listener Ratings of Similar Group Performances ...... 53

4.8 Means and Standard Deviations of Expert Listener Ratings of Different Group Performances ...... 54

4.9 Means and Standard Deviations of Expert Listener Ratings of Similar Female Performances ...... 55

4.10 Means and Standard Deviations of Expert Listener Ratings of Different Female Performances ...... 55

4.11 Means and Standard Deviations of Expert Listener Ratings of Similar Male Performances ...... 56

4.12 Means and Standard Deviations of Expert Listener Ratings of Different Male Performances ...... 56

vii

LIST OF FIGURES

3.1 High gestural condition ...... 30

3.2 Low gestural condition ...... 30

3.3 Singer participant wearing the head-mounted directional microphone ...... 33

3.4 Standing arrangement of singer participants ...... 34

3.5 Musical selection ...... 36

3.6 LTAS plot of group 4 singing on [dɔ] in high (red) and low (black) gesture conditions ... 37

3.7 LTAS plot of group 4 singing on [di] in high (red) and low (black) gesture conditions ... 38

3.8 LTAS plot of group 3 singing on [dɔ] in high (red) and low (black) gesture conditions ... 38

3.9 LTAS plot of group 3 singing on [di] in high (red) and low (black) gesture conditions ... 39

3.10 Rating scale for expert panel perceptual analysis ...... 41

4.1 Interaction of gesture height and gesture order in intonation ...... 45

4.2 Interaction of gesture height and gesture order in vibrato extent ...... 48

4.3 Interaction of gender and vowel in vibrato extent ...... 48

viii

ABSTRACT

In this study, I explored possible effects of gesture height, specifically the location of the horizontal plane, on singer performance. Research questions for the study included: 1) Do singers change intonation, vibrato rate, vibrato extent, vibrato duration, and tone quality in response to gesture? 2) What are singer participants’ perceptions of the effects of gesture on singing? 3) How do expert listeners rate individual and group performances in response to high and low gesture heights? Four stimulus videos were created that featured a conductor directing two verses of My Country, Tis of Thee. Two of the videos incorporated a high conducting plane, with the ictus between the clavicle and the chin of the conductor. The other two used a low conducting plane, with the ictus between the navel and the bottom of the rib cage. For each conducting height condition, the video included instructions for the participants to sing the excerpt on [di] or [dɔ]. Conductor facial expressions and cues were consistent between videos. Participants (N = 48) sang the musical excerpt in all 4 conditions, in a counterbalanced order. Participants were all at least 18 years of age, majoring in music, with voice lesson and choir participation history. Participants were recruited from intact choirs in the College of Music, and signed-up for appointments in groups of 8, two singers per part, SATB. Participants were fitted with a head-mounted microphone apparatus and placed in a semi-circle facing the video screen. Both individual and group audio recordings captured performances in response to all video conditions. Participants completed a post-participation questionnaire regarding their perceptions of gesture in a general sense and in the specific task of this study. From the group performances, recordings were chosen via LTAS plots for analysis by a panel of expert listeners (N = 11). The listeners rated two group recordings; one group evidenced similar responses to the two gestural conditions (according to the LTAS plots), and one group demonstrated different responses to the gestural conditions. Four individual recordings were also rated: (a) a female performance exhibiting similar responses to gesture according to intonation and spectral center of gravity measures, (b) a female performance exhibiting different responses, (c) a male performance exhibiting similar responses, and (d) a male performance exhibiting different responses. Acoustic analysis of individual performances revealed that singers sang slightly less flat in the high condition than the low condition, although there was an interaction with order: Participants sang less flat in response to the second gesture height they saw. No significant differences were seen in measures of vibrato rate or duration between gesture heights. Another order effect was evident in vibrato extent: singers reduced their vibrato extent in the second

ix condition, but by a greater magnitude in the high-low order than the low-high order. No significant differences occurred in tone quality (spectral center of gravity) measures between gesture conditions. Singers' perceptions of the effects of conductor gesture revealed that participants believe that gesture does affect singing, with dynamics, tone quality, and breath being the most frequently cited responses. The expert listener panel did not discriminate tone quality differences between gestural conditions of group performances or the individual female responses. Across the comparisons of male individual recordings, the expert panel rated the low gesture condition significantly higher in tone quality. The present study neither corroborates nor contradicts pedagogical assertions of the effect of gesture on intonation and tone quality. The subjective nature of tone quality, and the difficulty inherent in its quantification, warrants further research. Implications and ideas for future study are discussed.

CHAPTER ONE

INTRODUCTION

The moment a conductor steps onto the podium to give the first downbeat of a performance, a series of nonverbal interactions take place between conductor, performers, and audience members. To an outside observer, the field of choral conducting may appear as a series of random arm movements, facial expressions, and other mysterious gestures; these gesticulations somehow inspire a group of musicians to join their voices to a product whose outcome is greater than the sum of its parts. The Oxford Companion to Music defines a conductor as “the director of an ensemble, responsible for all aspects of performance” (“Conductor”, 2014). Interestingly, an alternate definition of the word, from Merriam-Webster Online is “a material or object that allows electricity or heat to move through it” (“Conductor”, 2014). The two meanings might be combined to view a conductor as a person who allows the score to flow through their gestures to elicit a response from the present musicians. Garretson (1981) states, “Your basic function as the conductor is to interpret the music for the singers… The conductor must be more than a mere time beater.” (p. 1) In her book The Modern Conductor, Green (1987) opens the first chapter, “So You Want to Be A Conductor” with these words:

To stand in front of an orchestra, band or chorus and beat time does not make one a conductor. But to bring forth thrilling music from a group of singers or players, to inspire them (through one’s own personal magnetism) to excel, to train them (through one’s own musicianship) to become musicians themselves, personally to feel the power of music so deeply that the audience is lifted to new heights emotionally, or gently persuaded, through music, to forget momentarily the dust of earth and to spend a little time in another world – yes, this can be called conducting. (p. 1)

This artful description calls upon the conductor to be knowledgeable in a number of capacities to prepare an ensemble for performance. The conductor’s success relies upon the ability to communicate that knowledge to the ensemble and the audience. While this communication often takes place through verbal relay, nonverbal communication plays a large role in the transmission of the conductor’s intent. The common condition of inadequate

1 rehearsal time in preparation for performance has spurred the need for conductors to use rehearsal time with increasing efficiency. Nonverbal communication can enhance, support, or entirely replace the spoken word, improving the pace and efficiency of rehearsal time. The crux of nonverbal communication in the rehearsal rests with the need for the nonverbal message to agree with the encoder’s intent. Kohut and Grant (1990) elaborate on this: “Do not waste time telling the performers to crescendo in bar two when your right hand beat pattern does not change in size. Do not yell at the group to sing softly when your beat size clearly indicates a forte dynamic level” (p. 103). According to Darrow and Johnson, (2009), “There is no substitute for having a quality message to convey; however, adequate nonverbal skills are necessary to aid in its delivery” (p. 277). Gordon (1989) posits “Today’s conductors have become far more sophisticated at guiding ensembles by use of a functional sign language which we call gesticulation” (p. 63). He further explains three functions of gesticulation: (a) accounting for time and place within music, (b) preparing performers for music action, and (c) coordinating and regulating technical aspects of performance, including dynamics, articulation and tempo. While the use of nonverbal communication in the rehearsal appears to be desirable, researchers have demonstrated that young conductors rarely succeed at accurately assessing their proportional use of verbal vs. nonverbal instructions. Napoles and Vasquez-Ramos (2013) explored perceptions of pre-service teachers on the amount of time they spent in teacher talk. Initial estimates by the participants after a short teaching episode grossly underestimated the amount of time spent talking. After watching a video of a teaching segment and taking data on teacher talk duration, participants reduced teacher talk by half and increased accuracy in duration estimation in subsequent teaching episodes. Eichenberger (1994) challenged his conducting students to reduce teacher talk by imposing the “7-word rule,” which mandates that the conductor utter no more than seven words when giving the choir an instruction. His instructional conducting video What they see is what you get (1994) goes further with the idea of reducing teacher talk and relying increasingly on nonverbal communication, or gesture. He makes several assertions based on his years of anecdotal experimentation in the choral rehearsal on the effect of gesture on choral sound. Garretson (1981) holds a similar value system on the use of verbalization in rehearsal, stating, “The value and importance of demonstrations and verbal explanations should not be minimized. During rehearsals, however, lengthy verbal explanations should be kept to a minimum, for through his technique the experienced conductor can more quickly help the group achieve the desired interpretation” (p. 7).

A considerable body of literature substantiates the use of sequential patterns in teaching, which help codify time use. Yarbrough and Madsen (1998) define the sequential pattern model of music instruction as (1) teacher presentation of a task, (2) student response, and (3) related and specific teacher reinforcement. The use of sequential patterns has been shown to enhance teacher effectiveness as well as improve student attitude, behavior, and musical performance (Arnold, 1991; Yarbrough & Price, 1981, 1989). The two teacher-directed tasks, presentation of a task and related and specific reinforcement, greatly impact the pacing of a rehearsal. Capable coordination of verbal and nonverbal communication can improve efficiency in these tasks. Sousa (1988), Mayne (1992), and Cofer (1994, 1998) all explored the idea of conducting gestures as nonverbal emblems. A gesture may be considered an emblem when intent matches perception by 70% of a population. In recent years, researchers have begun to explore empirically the effect of gesture on a variety of choral phenomena. Fuelberth (2003a, 2003b, 2004) explored varied left-hand conducting gestures on singer tension in a series of studies, finding that fisted, stabbing, and palm up gestures elicited the highest tension ratings. Manternach explored the effects of conductor preparatory gesture on singer upper body movement (2011), finding that choristers typically mimic the conductor’s body movement, and the effect of conductor lip rounding on singer lip and eyebrow postures (2012), finding a significant increase in singer lip rounding and raised eyebrows in response to conductor model. Daugherty and Brunkan (2013) also explored conductor lip rounding, adding acoustic measures of singer response, finding that 90% of singers rounded their lips in response to conductor modeling. Grady (2013) investigated traditional, lateral-only, and horizontal-only patterns on acoustical and perceptual measures of choral sound, finding that the vertical-only pattern was most in tune and most preferred by an expert listening panel. Brunkan (2013) studied the use of three different conductor and participant gestures (low circular gesture, high circular gesture, and traditional conducting pattern) on individual singers’ intonation and tone quality. The low circular gesture elicited the best intonation from participants and was perceived by an expert panel as the most balanced tone quality.

Need for the Study A major limitation of the studies heretofore mentioned includes the manufactured reality of the choral experience and the inability to measure individual singers while in an authentic choral setting. While LTAS (long term average spectrum) measures provide useful data for analysis of some aspects of choral performance, much detail vanishes in the process of averaging. In studies examining the solo singer in the choral context, participants often sing

3 along to a prerecorded choral excerpt, or are asked to respond to some other stimulus as if they were ensemble members. This study attempts to identify a method for isolating individual voices in an ensemble setting through the use of individual directional microphones. To date, there appears to be no research specifically exploring the effect of gesture height using a traditional conducting pattern on individual and group performance of singers.

Purpose Statement To that end, the purpose of this study was to explore the effects of gesture height, or the location of the horizontal conducting plane, on acoustical measures of individual and group performances in response to gesture, as well as expert listener perceptions of performed intonation, vibrato, tone quality, vocal tension, and dynamics. The following questions guided this investigation: (1) How do singers perform in response to gesture? Do singers change: (a) intonation (as measured by fundamental frequency), (b) vibrato rate (cycles per second), extent (average cents distance between peaks and troughs of the vibratory cyles) and duration (percentage of time a discernible vibratory cycle is present), and (c) tone quality (spectral center of gravity)? (2) What are singer participants’ perceptions of the effect of gesture on singing? (3) How do expert listeners rate individual and ensemble performances in response to higher and lower gestures in the following categories: intonation, vibrato, tone quality, vocal tension, and dynamics?

Delimitations While a major aim of the present study was to investigate individual singer response in a more realistic choral setting than previous research, this study also manufactured a choral experience. Participants were not members of an intact choir, and had no experience singing together prior to this study. Drawn from several existing ensembles, participants sang in groups of eight, a smaller ensemble than a typical choir. Participants did not sing with text, rather they were limited to the two syllables [di] and [dɔ]. Participants sang in unison in the appropriate octave for their gender, and watched a videotaped conductor. Participation was limited to those students at least 18 years of age, majoring in music, with prior voice study. Given these limitations, due caution is advised in making generalizations to an authentic college-age choral ensemble context.

CHAPTER TWO

REVIEW OF LITERATURE

The following review of literature serves to explore several topics central to the understanding of gestural language and its influence on singer performance, using both pedagogical texts and empirical research. Topics covered will include: (a) common gestural themes in conducting pedagogy texts, (b) verbal and nonverbal communication in music, (c) teacher talk and time use in rehearsal, (d) kinesics, or the use of nonlinguistic communication, and (e) conducting gesture. Subsequently, this review will explore topics connected to individual and ensemble musical performance, including: (a) intonation, (b) tone quality, (c) vibrato, (d) solo and choral singing modes, and (e) singer health.

Pedagogical Practices in Conducting Conducting pedagogues largely agree that the gestural language employed by conductors conveys decidedly more information than communication of time displayed in space. It has been suggested that the conductor serves as the conduit through whom the intentions of the composer travel to the performers (Decker & Kirk, 1988; Garretson, 1981; Gordon, 1989; Green, 1987). Nowak and Nowak (2002) conveyed this notion with the title of their text Conducting the Music, Not the Musicians. Conductors also agree that numerous aspects of gesture impact performer response (Gordon, 1989; Kaplan, 1985, Nowak & Nowak, 2002). Garretson stated, “Technique … refers to the multitude of devices used by a conductor to convey the intent of the music and to achieve musical and artistic results. Fundamental conducting patterns, important as they are, are really subservient to the bodily and facial expressions that reflect the mood of the music. Nevertheless, they provide a basis from which to start” (1981, p. 7).

Conductor Posture One of the first bodily behaviors evidenced in rehearsal and performance is posture. According to Lamb (1974), the posture of a conductor is equally as important as the singer’s posture. Garretson (1981) asserts that the conductor should resemble the “correct” posture of the singer. “One foot is placed several inches ahead of the other and the feet are spread apart so as to provide a solid foundation; this position allows the conductor to remain alert and responsive to the musical situation” (p. 21). Others propose that good posture in itself does not constitute a final goal, but that the conductor’s stance can function as effective and efficient communication (Demaree & Moses, 1995). Gordon (1989) includes in his text a detailed

5 synopsis of the proper postural positions for the feet (shoulder width apart), chest (comfortably high), alignment (straight torso without protruding chin or posterior), and suggests some direct effects on singer performance due to postural choices. For example, conducting with an open leg stance might suggest a robust sonority. Other pedagogues offer more practical motivations for attention to posture, stating that a wide stance might appear ungainly and undignified. Green (1987) wrote, “Show poise, dignity, and sincerity, and the audience will respond with appreciation and respect” (p. 13). Decker and Kirk (1988) state, “To be understood clearly, the conductor finds an erect posture and a commanding stance of primary importance. Choral sound is affected by a conductor’s body language. When the conductor’s posture is not erect, it cannot remind the singers to maintain the posture necessary to support the vocal sound” (p. 7).

Hand and Arm Position A practical extension of postural concerns for conductors involves the position of the hands and arms while preparing and performing conducting gestures. Some conducting texts include very specific instructions regarding hand and arm posture; for instance, the palm of the right hand should be turned down (Decker & Kirk, 1988; Demaree & Moses, 1995), the arms should be extended until the tips of the fingers are 12-16 inches from the body (Gordon, 1989), the arm should form no more than a 45-degree angle with the body (Decker & Kirk, 1988), and the arms and hands together with the back form a circle (Nowak & Nowak, 2002). Conversely, others either do not include any directives regarding hand and arm position, or offer a vague description dependent on the style of the music or ensemble (Lamb, 1974; Green, 1987).

Conducting Plane and Size of Beat The location of the conducting plane appears to elicit divergent opinions from master conductors. Some offer either no mention of conducting plane (Kaplan, 1985), or a generic description of the conducting plane with little critical language attached (Gordon, 1989; Garretson, 1981; Lamb, 1974). Lamb (1974) suggests that the arms be raised “high enough to be seen clearly by the ensemble and low enough to be comfortable – may seem vague, but it is true” (p. 117). He elaborates by stating that the conducting plane will vary according to the size of the ensemble, the distance and position in comparison to the conductor, and the conductor himself, but does offer that the median height should be just below the shoulder. Garretson (1981), while never giving a point of origin for a default conducting plane, offers both practical and musical reasons for adjusting gesture height, citing visibility in a large ensemble setting and the opportunity to physicalize a musical phrase with gesture height. Nowak and Nowak (2002)

6 simply state that the plane should be high enough to coordinate with eye contact, around elbow height. According to Gordon (1989), “Conducting a pattern close to the abdominal area will tend to produce a well-supported tone” (p. 92). Decker & Kirk (1988) state on several occasions that elbows are to remain close to the body, with the arm forming no more than a 45-degree angle to the body; these two directives serve to keep the conducting plane low. Other pedagogues offer in their textbooks and instructional materials very detailed instructions regarding the height of the conducting plane, including exercises for practice (Demaree & Moses, 1995; Eichenberger, 1994). Judging by the sheer amount of text given to the topic, Demaree & Moses consider the height of the conducting plane (referred to as the conducting frame in their text) to be a crucial aspect of conductor technique. A method for practicing gesture is outlined, incorporating a physical prompt for size and height of gesture. The authors urge students to cut out rectangles of various sizes, each corresponding to a dynamic level. These rectangles become tools for practicing beat patterns, using the outer edges as a guide for the baton. In addition, they specify a note regarding height of the frames: “If you must sometimes practice in front of a desk or a music stand, each beat should strike the surface, at least slightly. (Be certain the desk or stand is high enough, however, that you are not beating time at your waist.)” (p. 31).

Verbal and Nonverbal Communication in Music While much communication during rehearsal takes place verbally, communication between a conductor and ensemble during performance takes place almost exclusively in a nonverbal fashion. Through traditional and non-traditional conducting patterns, eye contact, facial expressions, and proximity, conductors transmit a wealth of information. Researchers have examined nonverbal communication in a myriad of ways. One aspect includes the establishment of rapport through nonverbal behaviors. In a non-music study, Vacharkulksemsuk and Frederickson (2012) examined how nonverbal interactions changed between pairs of strangers completing a self-disclosure task, in which they asked questions of each other, learning increasing personal information about each other, versus pairs of strangers completing a control task, in which they edited a scientific journal article. The pairs completing the self-disclosure began to exhibit shared movements, moving in synchrony with each other. A post-test questionnaire revealed higher feelings of rapport between the pairs who self-disclosed and moved in synchrony. Similarly, Darrow and Johnson (2009) examined perceived rapport of preservice music therapists and teachers through their nonverbal behaviors. They concluded

7 that eye contact and gestures were most important to rapport ratings; however, ratings were not dependent upon nonverbal factors alone, but also on verbal behaviors and skills.

Time Use in Rehearsals Time use of conductor verbal and nonverbal behaviors and how they function in rehearsals has been examined. Hicks (1975) wrote an anecdotal article titled “Would you enjoy one of your rehearsals? Knowing the four functions of rehearsal language will lead to better communication and a better performance.” He asserted that language comprises four functions in a rehearsal: (a) appraisal (evaluative), (b) apprisive (informative), (c) prescriptive (remediative), and (d) consummative (perfective). The idea of the functions of teacher talk led to a series of empirical studies on the use of time in rehearsals. In a descriptive study on the time use of high school choral directors, Watkins (1986) found that directors spent approximately equal amounts in student singing (50%) and teacher verbalization (50%). He found no statistical differences in teacher talk percentages related to teaching experience or concert proximity. The teacher verbalizations were subsequently categorized into three modes: (a) musical/technical language (40.3%), (b) modeling and musical/technical language (15.4%), and (c) modeling alone (12.4%). Goolsby (1996) examined time use of experienced, novice, and student teachers in instrumental rehearsals. He measured prep time, initial teacher talk, time in warmup, time during each musical selection, breaks, final teacher talk, and dismissal. Student teachers talked the most and allowed students to play the least. Experienced teachers exhibited more efficient use of time; they used the most nonverbal modeling, established on-task behavior the quickest, and talked the least during rehearsals. In a second study, Goolsby (1999) videotaped 10 expert teachers and 10 novice teachers preparing the same piece for adjudicated performance. Findings demonstrated a similar trend to the earlier study: novices used more time overall and spent more time in verbal instruction. Expert teachers spent a greater percentage of the rehearsals in performance than novices did. Novices frequently stopped and restarted the ensemble without providing feedback or instruction. Experts addressed balance, style, tone, and intonation more frequently than novice teachers. In a study of choral rehearsals, Napoles (2007) examined the effect of duration of teacher talk on attitude, attentiveness and performance achievement. Two choruses rehearsed two pieces under two conditions: low teacher talk (20% of rehearsal) for one piece, and high teacher talk (60% of rehearsal) for the other piece. She found that students were less attentive during teacher talk than during performance in both conditions. Surveys administered daily,

8 video observation, and expert listener ratings of performances (by 30 choral directors) illuminated a possible interactions with the two pieces. The advanced chorus preferred the piece rehearsed under the high teacher talk condition, and its performance was correspondingly rated higher. Beginning students liked both pieces the same, and their performance of the piece rehearsed under the low teacher talk condition was rated higher. Napoles and Vazquez-Ramos (2013) subsequently examined perceptions of preservice teachers’ time spent in teacher talk. They compared self-estimates, peer estimates, and actual time. Upon initial trial, preservice teachers vastly underestimated time spent in teacher talk. After viewing a recorded teaching episode of themselves and calculating time spent talking, teachers reduced teacher talk by half and could more accurately estimate time spent talking during rehearsal. While gestural language has been shown effective, researchers have also investigated the efficacy of verbal and nonverbal language in communication, both independently and in combination. The combination of verbal and nonverbal communication presents interesting implications. Researchers in psychology (Mehrabian & Ferris, 1967; Woodall & Burgeon, 1981) have indicated that when nonverbal behaviors appear in conflict with verbal behaviors, verbal communication might be ignored. Stauch (1986) found that verbal and nonverbal behaviors of selected collegiate choral directors generally matched. Skadsdem (1997) studied the effect of conductor verbalization, dynamic markings, conductor gesture, and choir dynamic level on the dynamic responses of participant singers. Participants (N = 144), while watching a videotaped conductor and hearing a choir through headphones, sang a musical selection while referring to a score. Verbal instructions from the conductor elicited a stronger dynamic response than any of the other conditions. Because the piece being performed was not memorized, singer focus of attention on the conducting gesture ranged from 44-65% of the time. As singer attention on the conductor increased, performances reflected more gestural intent than written instructions. Similarly, Napoles (2013) also explored the efficacy of verbal instructions and conducting gestures, focusing on word stress and articulation. High school choral students (N =44) performed Music Alone Shall Live in several conditions: with and without word stress, and with varied articulations, while watching a videotaped conductor and reading verbal instructions. She found that expert listeners (N = 30 experienced choral teachers) perceived more articulation and word stress as a result of verbal instruction. Congruency of gesture and verbalization also demonstrated important implications; more articulation and word stress were present when verbal directives and gestures agreed compared to when they were incongruent. Responses were stronger from both communication modes than from the verbal mode alone, but only when the

9 messages were the same. Napoles offered some implications of this research; while the results suggest that verbal directives appear to elicit stronger responses than nonverbal, there are occasions where verbal communication is impossible. In performance, conductors rely solely upon nonverbal communication. The implications for teacher education point to a need for clarity of verbal instruction, a focus on ensuring that verbal and gestural messages are congruent, and refining gestures to be more easily understood.

Kinesics While the above studies have demonstrated that congruency of verbal and nonverbal behaviors play a central role in communication, the following studies focus solely on nonverbal communication in music. Researchers have turned to kinesics, the study of the relationship between nonlinguistic body motions and communication, in order to explore the influence of body language on music performance. Recent researchers have examined the use of movement in perceptions of expressivity. Juchniewicz (2008) asked participants to rate phrasing, dynamics, rubato and overall performance of three conditions of a pianist’s performance to a previously recorded excerpt. While the audio remained the same, the pianist performed under three conditions: (a) no movement, (b) head and facial movement, and (c) full body movement. The more the pianist moved, the higher the ratings. Sasanfar (2012) assessed perceptions of expressivity in collaborative performance based on the body movement of the accompanist. This study incorporated both aurally expressive and unexpressive stimuli combined with visually expressive or unexpressive stimuli. Music majors were able to distinguish between the two aural conditions, while non-music majors were influenced to a greater degree by visual expressivity. Kotsonis (2014) explored audience perceptions of expressivity in choral performance with and without singer body movement. Participants rated movement as more appropriate for a Moses Hogan piece than for a Bach piece, but in both pieces, expressivity ratings were higher in the movement condition than in the still condition. Camurri, Mazzarino, Ricchetti, Timmers and Volpe (2004) found that communication of emotional intensity as displayed by pianists mainly takes place through tempo and dynamics, with performer movement reflecting these intensity levels. Movement has also been studied as a means of instruction in rehearsal. Chagnon (2001) observed three rehearsals each by five choral directors. The directors used movement to address musical qualities, vocal technique, choral technique (such as balance, blend, and diction), and interpretive qualities. Hibbard (1994) observed one master conductor’s use of movement in rehearsal, and categorized the movements using Laban Movement Analysis. Ten categories of

10 movements included: (a) differences in sound, awareness of changes in timbre, (b) breath support, energy, activity, (c) improving posture, stance, (d) locomotor movements, walking (d) movements for setting tempo, rhythm, (e) movements to sustain energy, phrasing emphasis, (f) movements imitating lifting of soft palate, tone quality, (g) other iconic, intrinsic movements for tone quality, intonation (h) movement for articulation, diction, attacks, releases, and cutoffs (i) sequences of movements with multiple purposes (j) incorporating movements into the conducting pattern. Hibbard offered five conditions for movement to be pedagogically effective, i.e. increasing time ‘on-task,’ and providing a nonverbal communication of the “ineffable, expressive qualities of music.” He also stated that a majority of movements have either a direct or indirect effect on tone quality. Atkins & Duke (2013) explored how four different pedagogical techniques might affect tone quality of 30 singers: (a) singing while feeling vibrations on the throat with the palm of one hand, (b) singing with the index and middle fingers placed on the cheeks on either side of the nose, commonly referred to as ‘aiming for the mask’, (c) directing sound to a microphone, and (d) directing the sound to a point on the wall across the room. Singers were also recorded in a baseline condition, with no focus instructions. Atkins, as an expert listener, ranked each excerpt from 1-5 (best to worst) for tone quality. The mask condition (fingers on either side of the nose) yielded the largest amount of first place rankings, followed by the microphone and the point across the room. The researchers also examined the acoustic data for harmonic-to-noise ratio, which will be discussed further under tone quality analysis.

Mimicry Mimicry and the role of mirror neurons in nonverbal communication have emerged as a strand of study in music education research. In non-music investigations, researchers observed that mimicry could be used as a form of communication; observers of an injury are more likely to flinch or wince when they make eye contact with the victim (Bavelas, Black, Lemery & Mullett, 1986). Conducting texts make assertions about the usefulness of mimicry and mirror neurons in rehearsal and performance. As early as 1918, Gehrkens stated “the conductor works largely through the instrumentality of instinctive imitation… his methods are founded upon the fact that human beings have an innate tendency to copy the actions of others, often without being conscious that they are doing so” (p. 3). Krone (1949) claimed that unconscious imitation could decrease the need for verbal directives from a conductor; i. e., singers will imitate the posture of a conductor. Garnett (2009) stated specifically that mirror neurons may influence chorister responses to conducting gestures. In an effort to investigate the role of mirror neurons

11 and mimicry in choral conducting, Daugherty and Brunkan (2013) explored the effect of conductor lip rounding on participant lip rounding. Singers (N =107) performed the first phrase of Ave Verum Corpus from memory while watching a videotaped conductor. Two videos displayed a life-size conductor on-screen in two conditions. In the first condition, the conductor maintained a neutral face, while in the second condition he rounded his lips. Analysis used a photo grid to measure participants’ horizontal and vertical lip distances during the [u] vowel. Results indicated that 90% of singers rounded their lips in response to conductor modeling as compared to baseline data. Manternach (2012) replicated and extended this study by exploring the effect of conductor lip rounding and eyebrow lifting on participants’ eyebrow and lip postures. Forty-seven singers sang the first phrase of Ave Verum Corpus while watching a videotaped conductor in two conditions. The conductor maintained a neutral facial expression in one condition, while in the second condition he modeled raised eyebrows and rounded lips on the [u] vowel. Manternach used motion capture technology to assess participants mimicking behaviors. Participants rounded their lips and raised their eyebrows significantly more when the conductor modeled these behaviors. Additionally, participants exhibited a residual effect, rounding their lips more than baseline on subsequent occurrences of the vowel when the conductor was no longer displaying rounded lips. Also of note, many participants noticed the lip rounding in the stimulus videos, but only a few noticed eyebrow lifting. Despite the lack of conscious observation of eyebrow lifting, mimicry still occurred.

Gesture As the conducting gesture is the primary means for communicating expressivity in performance, integrating non-traditional movement into the conducting gesture has been suggested for substantial consideration (Gambetta, 2003; Krudop, 2003; Wis, 1993). Gambetta (2003) explored incorporating principles of Laban Movement Analysis (LMA) with gesture. Four conductor participants conducted an excerpt of a Beethoven overture, and then received a 5-hour LMA training course. Two expert panels (one LMA, one conducting) watched both pre- and post-training videos. LMA experts confirmed that all four participants exhibited an expanded range of movement vocabulary in the post test, and the conducting experts concurred that the observed changes constituted a positive development for all participants. Krudop (2003) explored the use of kinesics in eliciting ensemble response, suggesting that the study of body movements, including hand gestures, facial expressions, or body language could elicit technical or emotional expressivity. This study incorporated a survey of music educators, observations of the application of kinesics by choral conductors, performance

12 evalutions of ensembles’ emotional response to conducting gestures using kinesics, and the creation of a unit of study for an undergraduate conducting course integrating kinesics in order to enhance nonverbal communication. A panel of experts indicated a strong preference for the presence of technical acuity in advance of emotion or expressivity in the rehearsal process. Wis (1993) observed two well-known conductors and their use of movement: Timothy Haskett and Rodney Eichenberger. She concluded that movement activities (a) allow for freedom and energy inherent in daily activities to be metaphorically conveyed to singing, (b) provide a visual accountability for individual learning, and (c) are less subject to misinterpretation than words, while capturing the indescribable qualities of music. Using a continuous response method, researchers attempted to define expressivity in the conducting gesture. According to one study, increased amplitude and speed of gesture conveyed higher levels of expressivity (Luck, Toiviainen, & Thompson, 2010). Visual stimuli have also been shown to influence aural perception. Schultz and Lipscomb (2007) asked auditors to identify the duration of several percussion notes performed with long and short gestures, finding that the gestures had a large impact on participants’ perceptions of duration. It follows, then, that conductor gesture has a large impact on ensemble evaluation (Gallops, 2005; Madsen, 2009; Morrison, Price, Geiger, & Cornacchio, 2009; Price & Mann, 2011; Price & Winter, 1991; Silvey, 2011, 2013). Several researchers have evaluated the effect of conductor’s gesture on performance ratings using identical audio tracks. Madsen (2009) found that poor conducting might have a greater impact on ensemble ratings than good conducting. Morrison et. al (2009) found significant effects of high and low expressivity conducting paired with identical audio stimulus. Ensemble ratings were higher during the high expressivity conducting condition. Price and Mann (2011) found significant differences in wind band performance evaluations; while watching seven different conductors paired with identical audio performances, evaluators gave performance ratings to both the ensemble and the conductor. Ratings were highly correlated (r = .97), and it appears that conductor performances account for more than 30% of the variability of the wind band performance ratings. A large portion of participants, who were free to comment on any aspect of the performances, commented positively on how well the performers followed the conductor. Given that the audio performances were identical between conductors, this makes a remarkable statement on conductor impact. Similarly, Silvey (2013) examined the role of conductor facial expressions in student evaluations of ensemble expressivity. Of the three conditions (approval, disapproval, and neutral), neutral facial expressions received the lowest rating. Reversing the variables of the

13 previous studies, ensemble performance also appears to influence ratings of conductor expressivity (Silvey, 2011). Intriguingly, written comments were directed to the conductor in the excellence performance condition, while they addressed the ensemble in the poor performance condition. In a study pairing perceptual ratings with practical ratings, Gallops (2005) found that musicians were able to realize expressive-interpretive gestures without any expressive markings in their music when responding to conductors who were rated as highly expressive.

Conducting Emblems The efficacy of conducting gestures greatly increases when the gestures have a universal meaning. Paul Roe, in his text Choral Music Education (1983), states, “Never stop an organization to say what can be shown with a gesture” (p. 260) In nonverbal communication, a gesture can be defined as an emblem when it is performed similarly and recognized by at least 70% of a given population (Knapp, Hall & Horgan, 2014). Sousa (1988) asked students to identify the intended meaning of 55 videotaped, common conducting gestures on a multiple- choice test. Overall, 38 of the 55 gestures could be considered musical conducting emblems; college students recognized 47, high school 37, and junior high only 19 of 55. Mayne (1992) continued this line of inquiry by adding facial expressions to the conducting gestures and methodology used by Sousa. Junior high students recognized 14 of 53 as emblems, high school students recognized 29, and college students recognized 40 of 53. Cofer found significant improvement in gestural recognition (1994) and performance (1998) on 20 of these conducting emblems following conducting instruction. The control group who took only the multiple choice test without any instruction recognized just 5 out of the 20 emblems. Consequently, conducting instruction appears to have value to young performing groups. Kelly (1997) explored this theory with a comparison of band performances by students who received conducting instruction with students who received no instruction. Of 10 beginning bands, 5 were given 10 minutes of conducting instruction for 10 weeks. Comparison of pre and post-tests revealed significant improvement in rhythmic performance, rhythm reading, and phrasing. No changes were observed in articulation, dynamics, or overall performance. The ability to exhibit conducting emblems is generally agreed upon as a necessary skill for young music teachers, so ostensibly a great deal of time is spent in undergraduate conducting courses on performance of different emblems. Byo (1990) explored whether independent observers could recognize intensity contrasts of beginning undergraduate conductors. A 77% correct response rate indicated that these young conductors were fairly successful at demonstrating high and low intensities.

Conducting instruction research has also investigated feedback methods for achieving desired hand positions. The use of a coin on the back of student hands showed no significant difference from verbal directions for maintaining a closed (palm down) hand position (Dease, 2007).

Conducting Gesture in Choral Performance Investigators have recently turned to the examination of the effect of nonverbal gestures on various aspects of choral sound (Brunkan, 2013; Manternach, 2011; Fuelberth, 2003b & 2004). In a series of studies investigating the effects of left hand gestures (the left hand traditionally demonstrates expressive intent by the conductor, while the right hand traditionally demonstrates the beat pattern) on perceived and performed tension in singers, Fuelberth (2003b & 2004) asked singers to watch a video of a conductor to evaluate whether varying gestures (which included (a) no change, (b) palm up, (c) palm down, (d) fisted gesture, (e) sideways, phrase shaping gesture, and (f) stabbing gesture) could cause vocal tension in a chorus. They identified fisting and stabbing as causing the most potential tension in one study (2003a), and in another (2004) the palm down gesture was perceived to cause more tension than the no change condition. She consequently tested singer response to the same gestures while performing a short song, using a panel of experts to evaluate tension as observed visually and aurally. Results indicated that fisted, stabbing, and palm up gestures elicited higher tension ratings than palm down, neutral, and sideways/phrase shaping gestures. Further, the sideways/phrase shaping gesture appears to stimulate lower tension than the no change condition. Vocal pedagogues give much credence to posture and head position in singer performance (Vennard, 1967; Miller, 1996). Therefore, Manternach (2011) examined the effect of conductor preparatory gesture on singer upper body movement. Singers (N = 60) sang a portion of My Country Tis of Thee while watching a videotaped conductor using five different preparatory gestures. Participants exhibited a trend of mirroring the conductor in head movement and shoulder movement. Simply, when the conductor moved his head up, participants moved their heads up, and likewise with shoulder movement. Less experienced singers exhibited greater shoulder movement than more experienced singers. Head and shoulder movement analysis incorporated a grid as a measurement tool. Brunkan (2012) examined the effects of watching and performing three types of gestures on singer tone quality and intonation. This three-phase study examined a traditional conducting pattern, a low circular gesture, and a high circular gesture. In the first phase (pretest), all singers

15 performed Happy Birthday in response to all three gestures. No differences were found between condition in this phase. In the second phase, singers were randomly assigned to one of two groups: (a) a control group who practiced the song with a metronome, and (b) an experimental group who practiced the piece while watching the conductor performing the gestures they had seen on the video in phase one, and (c) an experimental group who practiced the piece while watching and also performing the three gestures seen in phase one. In the third phase (posttest), singers performed the piece in all three conditions again while watching the conductor gesture. Results indicated that singers performed with significantly better intonation in the low, circular gesture condition, compared to a traditional pattern and high circular gestures. An expert panel of listeners also indicated that the most balanced tone quality occurred in conjunction with observation or performance of the low, circular gesture. The fact that no difference appeared between gestures in the pretest indicates that exposure to a gesture over time may be an important factor in gestural impact on singer performance.

Intonation A number of musical elements contribute to successful performance, and intonation plays a critical role in performance quality. Music educators devote great quantities of time to tuning (Witt, 1986), and most music performance assessment rubrics include intonation as a category. Furthermore, music researchers have devoted considerable study to the various aspects of intonation. Music listeners have demonstrated a preference for intonation over tone quality, and intonation errors in the sharp direction are preferable to intonation errors in the flat direction (Madsen & Geringer, 1976; Madsen & Geringer, 1981). Researchers have also examined the role of vibrato in perception of intonation. Geringer, MacLeod, & Allen (2010) asked 72 music majors (36 with string performance experience, and 36 without string experience) to match a non-vibrato stimulus tone to a pitch with vibrato. Means were similar for both groups, choosing a tone very near the mean frequency of the vibrato. String players decided quicker than the non-string group. Researchers have explored intonation performance practices of musicians with respect to three types of tuning historically used by Western classical musicians: (a) Equal tempered, (b) Just intonation, and (c) Pythagorean. Nickerson (1949) found that most string musicians tended toward Pythagorean tuning, in both solo and ensemble performance, although there was no strict conformity to one tuning or the other. While choral music accompanied by piano must be in accord with equal temperament, research in a capella singing has demonstrated that singers tend to veer away from equal tempered singing, toward just intonation (Howard 2007a) .

Howard (2007b) found that in a capella pieces with key modulations, this tendency for non- equal tempered tuning resulted in pitch drift as a necessity for staying “in-tune.” Duke (1985) examined wind players’ performances of several intervals, both ascending and descending, and found that the performers tended to underestimate size of intervals: ascending tones were played slightly flat, and descending tones were played slightly sharp. The magnitude of the changes were small and perhaps not musically meaningful. Van Besouw, Brereton, and Howard (2008) investigated the range of acceptable tuning (RAT) for synthesized tones with and without vibrato, finding that listeners were more forgiving of intonation errors in tones with vibrato by ten cents. Researchers and singing pedagogues have also investigated factors influencing singer intonation. In a best-practices style essay for the Music Educator’s Journal, Powell (1991) suggests that most intonation errors in singers result from technical inefficiencies: (a) jaw, neck, or shoulder tension, (b) poor posture, and (c) poor breath management. He offers three solutions in the article, included tuning lines vertically and horizontally, unifying vowels, and addressing vocal production flaws. Devaney (2011) examined intonation practices of singers in both solo and one voice per part SATB ensemble singing. She found more variability between groups than expected, but found differences in how singers approached ascending and descending intervals: half steps consisted of a smaller interval when descending than when ascending in pitch, and whole steps tended to be smaller when ascending than when descending. Jers and Ternström (2005) examined intonation of a 16 voice choir recorded individually while singing a Praetorius round in unison. During descending passages, singer intonation seemed to be very accurate relative to equal temperament. During ascending passages, however, the singers performed slightly sharp. Murbe, Friedemann, Hofmann, and Sundberg (2002) explored singers’ ability to match pitch with and without masking noise. They discovered that auditory feedback contributed substantially to singers’ pitch control. Ternström and Sundberg (1982) used contact microphones to measure intonation of singers singing either a third or a fifth above a stimulus tone. Stimulus tones were presented with and without vibrato, and with a variety of upper partials. Singers had the greatest ease of matching intonation when more of the harmonic spectrum was present in the stimulus tone, regardless of vibrato. Prame (1997) established that professional singers in recordings were sometimes ±40 cents or more from intended pitch. Since these recordings were commercially released, these deviations were either not noticed or judged acceptable. Seashore (1938) and Sundberg (1979) both noted that listeners seem generous and operate in an "interpretive mode" when listening to

17 singing; this tendency was recently called "vocal generosity" (Hutchins, Roquet, & Peretz, 2012). Hutchins et al. found a vocal generosity effect in both melodic and single tone conditions: Their musician and non-musician participants were 40% more likely to judge a sung note as in tune than its violin equivalent. Geringer, MacLeod, Madsen, and Napoles (in press) compared musicians’ discrimination of intonation in unaccompanied melodies performed by trumpet, violin, and voice and whether there were differences in melodies performed with and without vibrato. All three stimuli were perceived as more out of tune when there was no vibrato compared to vibrato. Melodies performed with vibrato were judged differently: voice examples were judged as more in tune than both instruments for both flat and sharp pitch deviations. Differences in perception between timbres may be influenced by characteristics of the vibrato itself such as modulation width, rate, and type. Geringer, MacLeod, and Sasanfar (2013) subsequently studied music majors' perception of intonation in accompanied performances. Listeners (N = 150) heard solo trumpet, violin, and soprano performances of Ave Maria (Bach/Gounod) accompanied by piano. Listeners were clearly more tolerant of pitch deviations in the vocalist's performance, and rated the singer more in-tune than the instrumentalists, particularly when deviations were in the sharp direction. Brunkan (2013), as discussed previously, found that a low circular gesture elicited more accurate intonation than a traditional conducting gesture or high circular gestures.

Tone Quality Analysis of tone quality in music relies primarily on subjective, perceptual judgments. Researchers have attempted to define the acoustical phenomena present during instances of “good” and “poor” tone quality, and variations in the harmonic spectrum are chiefly responsible for differences in varying singer tone qualities. Vennard (1967) explained the varying tone qualities available to the human voice through the Source-Filter Phenomenon. The oscillations of the vocal folds in the larynx (the source) create physical sound waves, which resonate in the pharyngeal cavities (the filter). Since humans can alter the shape of the resonating cavities by altering tongue, palate, and laryngeal placement, the voice has a great many tone qualities available. Bartholomew (1934) was among the first to attempt to define “good voice quality” in male voices, and described vibrato, intensity, and two aspects of tone quality: low formant and high formant. Low formant as described refers to a strengthening of the 2nd partial in the harmonic spectrum in “good” voices; “bad” voices tended to strengthen the 3rd, 4th, or 5th partials. The high formant as used by Bartholomew does not refer to what is now known as the

Singer’s Formant, a strengthening of partials in the 2400-3200 Hz range. Titze, Bergan, Hunter, and Story (2003) explored what filter adjustments singers make to achieve the vocal effects “twang” and “yawn.” Twang corresponded with a shorter vocal tract, narrower pharynx, and increased closed quotient, and the opposite for the perceptual evaluation of yawn. Other researchers have explored tone quality distinctions resulting from differing modes of singing, namely choral and solo. While strengthening the singer’s formant in solo western classical singing serves as a primary goal of voice teachers and students, auditors appear to prefer choral singing without a strong singer’s formant, especially when auditors have choral training (Ford, 2003). These sometimes divergent goals appear to cause stress for young singers; Detwiler (2009) found that singers with well developed upper partials (singer’s formant) diminished their vocal efficiency in the choral setting in order to blend with surrounding singers, and asserted that choral singing is not a “one size fits all.” Carter (2007) measured tone quality in 9 male singers (3 undergraduate underclassmen, 3 undergraduate upperclassmen, and 3 graduate students) in both choral and solo passages. Singers with less training demonstrated a larger difference in the use of singer’s formant between solo and choral singing (more in solo, less in choral). As training increased, the difference in the amount used between modes became insignificant, though the small sample size of this study limits the generalization of these results. Fagnan (2005) argued that the use of three bel canto techniques, (a) coup de glotte, (b) chiaroscuro, and (c) messa di voce, could enhance acoustical elements of choral singing. The coup de glotte technique strengthened spectral energy in the singer’s formant region, which may not always be preferred by choral directors. Chiaroscuro, the balance of bright and dark timbres in vocal production, improved intonation and allowed for a more homogeneous sound. Expert evaluations of vocal performances have been found to be somewhat unreliable. Wapnick and Ekholm (1997) assessed 21 expert evaluations of 25 excerpts of vocal performances. These excerpts included 17 recorded for the purposes of their study, two commercial recordings, and 6 randomly chosen repetitions of the original recordings to measure intrajudge reliability. Interjudge reliability increased dramatically with panel size, with a 10 judge panel yielding a Pearson coefficient of .90. Intrajudge reliability, which measured the consistency of individual judges on repeated excerpts, varied from .33 to .89, with a mean of .70, and did not demonstrate a significant correlation with age, teaching experience, or adjudication experience. Among the categories of evaluation, overall score and intonation accuracy yielded the highest interjudge correlation (r = .64 each), followed by freedom throughout vocal range, appropriate vibrato, evenness of registration, and resonance/ring. Color/warmth (r = .43) and

19 diction (r =.34) exhibited the lowest interjudge reliability. This low interjudge reliability of the color/warmth category, which addresses an aspect of tone quality, demonstrates the inherent difficulty in the task of assessing tone quality among singers. Ekholm, Papagiannis, & Chagnon (1998) compared expert ratings with acoustical analysis of spectral measures and vibrato rate, extent, and onset of 16 male singers (n = 4 countertenors, n = 7 tenors, and n = 5 baritones) performing a Mozart aria in a key of their choice. The panel of experts (n =7 expert voice teachers) evaluated performances for (a) color/warmth, (b) resonance/ring, (c) appropriate vibrato, and (d) clarity/focus. Ratings for color/warmth correlated highly with clarity/focus, and color/warmth with appropriate vibrato. Delayed onset of vibrato (defined as greater than 0.5 seconds) had the greatest effect on appropriate vibrato scores. Higher spectral energy in the singer’s formant range held a strong positive correlation with color/warmth ratings. Examples with high levels of noise components (elevated levels of nonharmonic spectral energy, often perceived as “breathiness”) received lower ratings in the clarity/focus category. Researchers have continued to seek methods to analyze tone quality in a quantitative manner. Three such methods will be discussed below: (a) the Singing Power Ratio, (b) the Harmonic to Noise Ratio, and (c) the spectral center of gravity. Omori, Kacker, Carroll, Riley, and Blaugrund (1996) introduced the Singing Power Ratio (SPR) as a novel means of examining spectral data. SPR calculates a ratio between the intensity of the peak harmonic between 2 and 4 kHz and the intensity of the peak harmonic between 0 and 2 kHz. The 2-4 kHz range corresponds with the Singer’s Formant, and it was argued that the SPR provides usable data even with soprano voices, who do not exhibit a singer’s formant. Lower SPR values (i.e. a smaller difference between intensity in the two spectral areas) indicate increased spectral energy in higher harmonics, which corresponds to perceptual measures of ring, or richness in the voice. Watts, Barnes-Burroughs, Estis, and Blanton (2006) also investigated the SPR as a measure for assessing differences between untrained talented and non-talented singers. Significant differences between groups (lower SPR values in the untrained talented singers and high SPR values in the non-talented singers) indicated that vocal tract resonance as measured by the SPR is an important variable in the perception of singing talent. The Singing Power Ratio appears to show promise in aiding discrimination between untrained singers. Atkins & Duke (2013) used two empirical methods for quantifying tone quality, the harmonic-to-noise ratio and formant distance to explore how four different pedagogical techniques might affect tone quality of 30 singers, as discussed earlier. The researchers examined the acoustic data for harmonic-to-noise ratio, which analyzes the ratio of periodic

20 components to aperiodic components (perceived as breathiness), and for formant frequencies. The harmonic-to-noise ratio did not prove to be a solitary indicator of good tone – only 11 of 30 (37%) participants had the highest harmonic-to-noise ratio in their best excerpt. Nineteen (63%) singers performed with the highest harmonic-to-noise ratio in a condition that was not ranked as their best. The researchers also analyzed formant frequencies in an attempt to define the perceived differences in “ring” or “resonance” between conditions. Statistical analysis revealed no significant differences in the distances between formants, though expert listeners perceived changes in tone quality. This use of harmonic to noise ratio and formant distance in analysis of tone quality does not appear to reflect subtleties noted by the human ear. Another means of empirical examination of tone quality is the spectral center of gravity (SPG), sometimes known as spectral centroid. The SPG uses a single number to represent the spectrum, as measured by the average of frequencies present in a spectrum, weighted by their intensity. This measure appears to have a robust connect to listener perceptions of brightness. Grey and Gordon (1978) compared perceptual ratings of 16 woodwind, brass, and stringed instruments. Listeners heard pairs of instruments, either unaltered, or with spectral envelopes exchanged. For instance, in an exchanged pair, listeners heard the onset of a trombone tone with the body of a trumpet tone. Grey and Gordon used both a median representation of the spectrum and a centroid (mean) representation, and found that the centroid represented the spectral distributions more accurately than the median. Listeners heard very strong differences in perceived attributes of tone with the exchanged pairs. Schubert, Wolfe, and Tarnopolsky (2004) also examined spectral centroid as a means of quantifying timbre. They were interested in exploring the differences in perceived brightness of instruments playing in various octaves. The working hypothesis was that the same instrument would be perceived as brighter when playing a higher octave, so they compared listener perceptions with two measures: (a) spectral centroid and (b) spectral centroid divided by the fundamental frequency. They found that perceived tone quality better correlated with the spectral centroid rather than the ratio of spectral centroid to the fundamental. Caclin, McAdams, Smith, and Winsberg (2005) ran a confirmatory study on the efficacy of spectral centroid as a measure of tone quality. They concluded that spectral centroid functions as a major determinant of timbre, as reported by Grey (1997), Krimphoff, McAdams, and Winsberg (1994).

Vibrato As evidenced in Ekholm et al.’s (1998) research, the use of vibrato affects expert ratings of tone quality. Bartholomew (1934), Seashore (1938), Miller (1986), Prame (1997), Nair (1999),

21 and Bretos and Sundberg (2003) described parameters of performance practice regarding the rate and extent of vibrato in singers. Bartholomew identified a vibrato rate between 6 and 7 Hz. Seashore proposed an extent of 100 cents and mean rate of 6.5 cycles per second. Miller asserted that vibrato rates should fall around 6 cycles per second. Nair offered a wider range of acceptable rates: 4.5 – 6.5 cycles per second. Bretos and Sundberg analyzed ten recordings of famous sopranos for vibrato rate (M = 6.25Hz), extent, and intonation over time. Findings indicated that vibrato rate, extent, and intonation all changed over time on long sustained tones. Prame, in his examination of 10 professional recordings, found a wide scope of performance practices with regard to extent, ranging from ± 34 cents to ± 123 cents, with a mean of ± 71 cents. Diaz and Rothman (2003) analyzed vibrato samples classified by experts as “good” or “poor,” finding that periodicity of extent (consistency of vibrato width) appeared to be the greatest indicator of “good” vibrato. Ekholm et. al (1998), reviewed above, compared expert ratings with acoustical data from 16 singers. Vibrato characteristics had great impact on expert ratings. The expert panel scored singers exhibiting delayed onset as short as .5 seconds notably lower in ratings of appropriate vibrato. Delayed vibrato also influenced ratings in other categories, such as color/warmth. A balance of vibrato rate and extent also appeared to be desirable by voice teachers: singers with a large extent and low rate received lower ratings in appropriate vibrato, though the researchers stated that this aspect warrants more research. Researchers have likewise focused on singer vibrato in a pedagogical context. Mitchell and Kenny (2010) investigated the development of vibrato rate and extent of undergraduates throughout their degree program. With training, the mean vibrato rate decreased from a mean score of 5.38Hz in semester one to 5.21Hz in semester four, and the mean vibrato extent increased from .37 semitones to .48 semitones. Dromey, Carter, and Hopkins (2003) explored acoustical changes as singers attempted to match their vibrato rate to a stimulus tone; slower rates were associated with lower intensity and less steady vibrato, and faster rates were associated with increased intensity in the head register and increased vibrato extent in the chest register. Singers also demonstrated the ability to adjust their vibrato rate to match a stimulus tone. Continuing this line of inquiry, the researchers defined techniques used to change vibrato rate while matching a stimulus tone (Carter, Hopkins, & Dromey, 2010). Placement and breath energy were the most notable techniques employed; a fast vibrato becomes slower with diminished breath energy and more backwards placement, and vice versa. Weber (1992) looked at changes in intensity levels of varying pitches with and without vibrato. Participants sang vibrato tones with significantly more intensity than straight tones at the high pitch level.

Mitchell & Kenny (2004) explored the effects of “open-throat” technique on vibrato rate, extent, and onset. Six singers were recorded singing in three conditions as verified by a pedagogue present at the recording sessions: (a) optimal, using open throat technique, (b) sub-optimal, and (c) loud suboptimal. There were no differences between the suboptimal conditions in any of the vibrato measures, and no differences appeared between rates in any of the conditions. However, the optimal condition varied from the other two in both extent and onset measures. On average, singers reduced vibrato extent by 40% from the optimal condition to the sub-optimal and loud sub-optimal. Both sub-optimal conditions also demonstrated a marked delay in vibrato onset, which is associated with poorer overall vibrato according to the literature reviewed in their study. As vibrato traits can change with training, so too can they change depending on style. Even within the realm of Western Classical solo singing, Miller (1997) identified differences in vibrato between the national schools. Singers from the French school tend toward a fast vibrato, Italian and English lean toward a medium rate, and the German school favors a slow vibrato rate. Howard, Daffern & Brereton (2011) examined vibrato rate, extent, and duration of a soprano singing early music in three different performance styles: (a) early music mainstream, (b) operatic, and (c) clear smooth sweet chaste, a label given to describe a tone with very little vibrato. Rate remained constant between the operatic and early music mainstream, while extents and durations varied between all three styles. As expected, the operatic style had the largest extents and durations of the three. Vibrato has risen to national prominence recently due to its use, or non-use, in the choral experience. McCoy (2011) in an article titled “The Choir Issue” in the Journal of Singing, addressed three commonly debated concerns of voice teachers on the choral experience: vibrato, pure vowels in high tessitura, and rehearsal length. Regarding vibrato, he asserts (based on some casual experimentation with his vocal pedagogy students) that singers employ three methods to remove vibrato from a tone: they (a) increase their closed quotient by 10% or more, (closed quotient refers to the amount of time during phonation that the vocal folds are adducted, or closed), resulting in a tense tone, (b) decrease their closed quotient by 10% or more, resulting in a breathier tone, and (c) maintain an essentially constant closed quotient between tones with and without vibrato. The third group reports less fatigue during non-vibrato singing. McCoy recommends that teachers aim for balanced, easy breath support, laryngeal freedom, muscular release in the articulators (especially the tongue and jaw), and the willingness to make a different sound quality. Olson (2008) discussed the role of vibrato in the choral rehearsal from the standpoint of

23 a solo vocal pedagogue. She encourages open lines of communication between choral directors and voice teachers regarding students’ individual developmental needs, avoiding participation in choirs that demand exclusively straight tone singing, and for the solo singer “to understand his own voice, and to exercise awareness while singing in a choral ensemble.” (p. 564). The above articles describe perceptions of vocal pedagogues and choral conductors, but little empirical study has examined changes made by singers when performing in different modes. Mann (in press) explored the effect of singing mode on vibrato rate, extent, and duration. Thirty female music majors (n =15 choral music education majors and n =15 voice performance majors) were recorded individually singing both the soprano solo and the soprano line from the choral section of two pieces of music with identical solo and choral material. Singers performed with accompaniment presented through headphones, including a choir in the choral mode. Results revealed significant differences between modes in all three vibrato measures: faster rate, wider extent, and longer duration of vibrato in solo compared to choral mode in both pieces. No significant difference appeared between choral music education majors and voice performance majors. Ford (2001) discussed implications for conducting gesture on choir sound, citing one gesture in particular that could affect vibrato performance. An example of an iconic gesture might be used if someone were describing the smooth surface of the sheets on a neatly made bed. While describing the surface, the person might produce a smoothing motion with a flat hand moving back and forth from left to right on a parallel plane to the ground. The conductor, to smooth out overly vibrato-laden passages in music, could use this same motion” (p. 22). Mann (2013) explored this notion that gesture might impact vibrato. Thirty-two singers, performing as SATB octets, were recorded using contact microphones while watching a videotaped conductor. The conductor displayed two gestural conditions: (a) a primarily horizontal conducting pattern, with very small rebound, and (b) a primarily vertical conducting pattern. The conductor changed conditions midway through the musical excerpt. Two groups were asked to pay careful attention to gesture, and two groups were not given any instructions regarding the conductor. The groups who saw the vertical gesture first had a larger vibrato extent than the groups who saw the horizontal gesture first. However, a three way interaction occurred between order, gesture condition, and instruction group: the participants instructed to pay attention to gesture exhibited smaller mean extents in response to the horizontal conducting, only when the horizontal gesture was first. Participants to whom nothing was

24 mentioned regarding gesture established an extent based on the first gesture seen, and exhibited no apparent change in response to the altered gesture. When the first gesture was horizontal, the smaller extent (M = 66 cents) was maintained even when the conductor switched to vertical gesture. When the vertical gesture was first, the singers also maintained a larger extent (M = 108 cents), even when seeing horizontal gestures. This result seems to support the idea that gestures, among other nonverbal behaviors, may need to be verbalized as a focus of singer attention. Participants performed while reading

Solo and Choral Singing Modes The discussion amongst vocal pedagogues and choral conductors continues in regard to other aspects beyond vibrato. While vibrato remains a prevalent issue, there are many other factors to consider for young singers engaging in both modes of singing. The National Association of Teachers of Singing (NATS) and the American Choral Directors’ Association (ACDA) partnered for a series of roundtable discussions at both organizations’ national conventions. In a resulting article, Hansen, Henderson, McCoy, Simonson, and Smith (2011) discussed the most prevalent issues from both points of view: (a) tone quality, (b)soloist selection, (c) age/size appropriate repertoire, (d), voice placement, (e) rehearsal techniques, (f) vocal fatigue, (g) terminology, (h) needless repetition, and (i) outside activities. In a descriptive study involving a survey of voice students, voice teachers, and choral directors from colleges and universities in Ohio, Slusher (1991) identified some divergent teaching practices and expectations of voice teachers and choral directors. He found conflict between studio teaching and choral rehearsal expectations, as well as differences in intensity levels; i.e., whether a forte dynamic from the studio would be accepted in the choral rehearsal. Detwiler (2008) explored the challenges faced by undergraduate performance majors alternating between solo and choral singing. She found that singers with developed upper partials required a diminished efficiency in order to achieve blend with surrounding singers in choir, and suggested that choral singing is not a ‘one size fits all’ phenomenon. She offered a methodology based on a review of pertinent research for voice teachers and choral directors to address the sometimes divergent goals of the two settings. Reid, Davis, Oates, Carbera, Ternström, Black, and Chapman (2007) explored the acoustic characteristics of professional opera singers performing in opera chorus versus solo mode. They found little or no difference between modes in measures of vibrato rate and extent and presence of singer’s formant. Rossing, Sundberg and Ternström (1986) conducted an acoustic comparison of solo and choral singing by examining spectral data, intensity levels, and

25 articulation of 8 bass/baritones. Participants generally incorporated more singer’s formant power in solo mode and more fundamental power in choral mode, as measured by sound pressure level. In choral singing, singers adjusted their loudness to the levels they heard from other singers, while in solo mode, loudness depended much less on accompaniment. Differences were also observed between modes in articulation. Goodwin (1980) observed individual voices singing soloistically and blending with a prerecorded ensemble presented through headphones and found that blended sounds produced a stronger fundamental pitch with fewer and weaker upper partials.

Singer Health The main concerns vocal pedagogues have with some choral experiences center on singer health (Olson, 2008). Because the body is the instrument for singers, vocal health and hygiene are critical factors in successful performance. After surveying amateur singers on their perceptions of singer health at the World Choir Games, Kirsh, van Leer, Phero, Xie, and Khosia (2013) found a moderate correlation between suboptimal singing behaviors and vocal fatigue. Singing outside one’s comfortable pitch range revealed the strongest association with vocal fatigue, while singing too loudly generated only mild association with fatigue. Regarding pure vowels in a high tessitura, McCoy (2011) encourages conductors to understand that above G5, pure vowels are acoustically impossible, based on the formant structure of vowels. Asking sopranos to produce pure vowels in this tessitura may result in a struggle that could increase vocal tension, leading to fatigue and potential injury. Instead, he encourages conductors to allow singers in extreme upper ranges to modify vowels toward [a]. Cook-Koenig (1995) interviewed prominent choral directors regarding their opinions of causes and suggestions for prevention of vocal fatigue. Several conductors identified conductor tension as a factor. She quoted Harold Decker, “the choral conductor influences good singing responses through his gestures. If he is tense in his physical movement or he/she has poor body stance, it will be reflected in the singers. If he encourages the flow and grace of sound in his/her conducting movements, the result is more than likely to be freer and less tense.” (p. 83). The Lombard Effect, the involuntary tendency for speakers to increase their speaking volume in the presence of loud ambient noise, also appears to play a role in singing. Tonkinson (1994) asked singers to perform The Star-Spangled Banner along with a choir presented through headphones. The stimulus recording became progressively louder. After a pre-test recording, singers were asked to resist the urge to sing louder when the stimulus choir gained

26 intensity. Post-test measures indicated that singers can resist the Lombard Effect when instructed to do so. The Lombard effect in singing appears to be associated with choral sound. Murbe et al. demonstrated that auditory feedback functions as a crucial aspect of intonation accuracy. Ternström (1993, 1999) explored singer preferences of hearing their own voice compared to the reference sound of a choir. Singers preferred their own voice an average of 3.9 dB greater than the reference; values ranged between +1.5 - +7.3 dB. This preference has been termed the Self to Other ratio. Daugherty (2001) lists this ratio as an important factor contributing to overall choir sound, along with room acoustics, formation, and spacing. Daugherty’s (2013) extensive research on spacing has demonstrated that lateral spacing (averaging 18 inches between singers in rows) and circumambient spacing (including columnar spacing in addition to row spacing) improved acoustical and perceptual measures of conglomerate choral sound.

Summary According to the literature, the conductor plays a vital role in the success of musical performance. Conducting texts make a case for the influence of gesture on sound, though they do not always agree on the manner and results of those influences. Nonverbal research indicates that rehearsal efficiency increases with effective nonverbal communication. The lip rounding studies of Manternach, Daugherty, and Brunkan indicated that some forms of nonverbal communication may produce a desired effect even when unaccompanied by a verbal directive. While researchers have established certain parameters of sound to be desirable by experts in the field, the research literature has yet to establish how specific conducting gestures translate to particular sounds in the choral discipline. Given the prominence of gesture in the rehearsal and performance processes of choral music, additional knowledge about the effects of gesture on sound would seem useful to choral conductors, singers, and pedagogues.

CHAPTER THREE

METHOD

The purpose of the present study was to explore the effect of conductor’s gesture height on singer response, as evaluated empirically and perceptually. Specifically, this study examined intonation, vibrato, and tone quality of singers in response to high and low conducting planes. Prior to commencing the study, approval from the FSU Human Subjects Committee at the Office of the Vice President for Research at the Florida State University was obtained (see Appendix A). This study was conducted in three phases: (a) preparing the stimulus video, (b) participant performance, and (c) performance evaluation.

Phase One: Preparing the Stimulus Videos Conductor Selection A local middle school choral director served as the conductor for the stimulus videos. The conductor was male, age 25, with 4 years conducting/teaching experience, and 18 years choral singing experience. He holds both the bachelor’s and master’s degrees in Choral Music Education. This teacher had been in the county for less than five years, making it unlikely for the singer participants to know this director. This served as an element of control for the possible confounding influence of previous experience with this director.

Independent Variables: Gesture Height and Sung Vowels Due to a lack of consensus among pedagogical texts and limited empirical study on gesture height, the present study was designed to explore two examples of gesture height. Anecdotal claims by pedagogues (Gordon, 1989; Eichenberger, 1994) that conducting near the abdomen creates a well-supported tone quality inspired the two gesture heights for this study. The low conducting plane condition used an ictus point between the bottom of the rib cage and the navel, and the high conducting plane used an ictus point between the clavicle and the bottom of the chin. Because tone quality is dependent upon vowel shape, this study incorporated both a forward and back vowel, defined as such by the location of the highest point of the tongue. The [i] vowel uses the most forward tongue position, while the [ɔ] vowel uses a tongue position much farther back. These vowels are sometimes referred to respectively as lateral and rounded vowels, due to the shape of the mouth opening in their formations. The average formant frequencies of

28 formants in the vowel spectrum vary according to vowel (Davids & LaTour, 2012), explaining the variety of tone quality possibilities between vowels.

Musical Selection My Country Tis of Thee was selected as the song to be performed for this research study. A number of factors influenced the selection of this piece: the familiarity of this song allows for minimal rehearsal time for participants, the range of the melody covers a minor seventh, and the slow tempo provides several notes long enough for analysis. Additionally, the familiarity of the melody to potential participants would not require use of sheet music during the recording task. Previous studies (Skadsdem, 1997; Mann, 2014) indicated that participants were less likely to pay attention to nonverbal communication from the conductor while holding sheet music.

Creating the Stimuli The stimulus videos were recorded with a SONY HDR‐PJ650V high-definition video camera. The conductor performed a traditional ¾ time conducting pattern in two gesture height conditions: (a) low conducting plane and (b) high conducting plane. The low conducting plane used an ictus point between the bottom of the rib cage and the navel, and the high conducting plane used an ictus point between the clavicle and the chin. (The ictus is the point at which the gesture changes direction, indicating the “beat” in the pattern.) In order for the conductor to maintain the two conditions, he performed the patterns while viewing his reflection in a mirror. Two squares drawn on the mirror with a nonpermanent marker delineated the two conditions and facilitated consistency by the conductor. The conductor used preparatory and cut-off gestures, as well as phrase shape indicators in the left hand. Additionally, he maintained a neutral but pleasant facial expression. Silvey (2013) found that ratings of ensemble expressivity ratings based on conductor facial expression were lowest for the neutral condition, compared with approving and disapproving facial expressions. A previous study by Mann (2014) indicated that participants ignored a stimulus conductor who gave no musical information beyond beating time. Participants also indicated in a questionnaire that a strictly neutral face was off-putting. A panel of experts, comprised of three graduate choral conducting students, analyzed the stimulus videotapes, and agreed that the preparatory, cut-off, and phrase shaping gestures were very similar and consistent between conditions. Figures 3.1 and 3.2 display still photos of the two gestural conditions.

Figure 3.1 High gestural condition

Figure 3.2 Low gestural condition

Recorded in a separate audio file, the conductor gave vowel instructions that were later edited into the videos for additional conducting conditions. Two vowel conditions were included per conducting plane: the [i] vowel and the [ɔ] vowel. The duration of each video was one minute, thirty-four seconds. In total, four stimulus videos were prepared for this study: (a) high conducting plane, [i] vowel, (b) low conducting plane, [i] vowel, (c) high conducting plane, [ɔ] vowel, and (d) low conducting plane, [ɔ] vowel. An initial consonant of [d] was chosen to avoid tension associated with a glottal onset, and because the voiced consonant [d] would not add substantial noise to the sound, a potential issue in acoustic analysis. At the beginning of each stimulus video, instructions were given against a blank screen: “Please sing two verses of the melody of My Country Tis of Thee on [di]. Here is your starting pitch:” Starting pitches were edited into the four videos using piano timbre generated with Garage Band. The key of F was selected for this study to provide a comfortable range while approaching the secondo passaggio (Vennard, 1967). Vocal tension resulting from conductor gesture would likely appear near the passaggio, so this key provided opportunities for analysis in both a comfortable range and in the passaggio for most singers.

Phase Two: Participant Performance Participants The second phase of this study was designed to gather data on singer response to gesture height. Singer participants (N = 54) were recruited from existing choral ensembles at Florida State University in Tallahassee, Florida. All participants met the criteria for inclusion in this phase of study: undergraduate music majors taking applied voice lessons at the principal level or above, and at least 18 years of age. I recruited participants from choral ensembles in the College of Music by asking for volunteers. Participants signed up for pre-arranged appointment times. Although 56 participants signed up for appointments to sing in octets, two singers failed to appear for their appointed time. In order to maintain equal size of eight singers, a research assistant sang in place of these missing singers, but was not recorded individually. In six instances, singers exhibited symptoms of a cold or some other vocal stress that made reliable analysis questionable, and these singers’ data were therefore excluded from analysis. The remaining 48 were equally divided by gender (n =24 males, n = 24 females). Voice parts consisted of soprano (n =11, 23%), alto (n =13, 27%), tenor (n =11, 23 %), and bass (n =13, 27%). Participants ranged in age from 18 to 24, with a mean age of 20.3; nine participants (19%) were freshmen, 10 participants (21%) were sophomores, 22 participants (46%) were juniors, and

31 seven participants (14%) were seniors. Participants majored in choral music education (n = 21, 44%), voice performance (n = 8, 16%), Bachelor of Arts in music (n =8, 17%), music therapy (n = 5, 10%), commercial music (n = 4, 8%), trumpet performance (n = 1, 2%), and undecided (n = 1, 2%). Years of voice lessons ranged from 1 to 11, with a mean of 4.6 years, and a standard deviation of 2.6 years. Years of choral experience ranged from 2 to 17, with a mean of 8.7 years, and a standard deviation of 3.4 years. Participants completed an informed consent form, approved by the university’s institutional review board (See Appendix A). The consent form stated (a) the study would explore current performance practices in choral singing, and (b) participation was voluntary without any compensation, and the choice to withdraw consent was available at all times.

Phase Two Recording Venue The recording task took place in the main choir rehearsal room at the university, 30 ft. by 43 ft. The room includes 6 built-in seated risers. Participants were all accustomed to singing in this space, as they were all recruited from intact university choral ensembles who regularly rehearse in this room.

Phase Two Equipment The head-mounted recording apparatus used for this study was designed specifically for this research study. Gentex Uni-Directional Electret Microphone Elements N100833-00 (Model 3107-0; See Appendix B for technical data) were mounted on a headset designed to quickly prepare singers for recording. The headset used clear safety goggles as the mounting system, which fit comfortably over glasses for any participants wearing prescription lenses (See Figure 3.3). Each microphone was mounted slightly to the corner of the mouth (within one inch) to reduce breath and articulation noise in the signal, while allowing for a strong directional signal from the closely-mounted microphone. Microphones were connected to a power supply unit (that provided 5V) and then connected to an 8-channel ZOOM R-16 Recorder/Interface/Controller. The eight individual monaural tracks were recorded using a 44.1KHz sampling rate with 24 bit resolution. Additionally, Gauge ECM-84 small diaphragm condenser microphones were placed in an "x-y" formation in front of the choir room and recorded the ensemble performances of the octets. These microphones were connected to an M- Audio Firewire 410 Interface to Pro Tools 10.2 on a MacBook Pro.

Figure 3.3. Singer participant wearing the head-mounted directional microphone.

Phase Two Pilot Testing A series of pilot tests was completed to determine syllables, placement of microphone on the face, and spacing of the participants on the risers in order to ensure the isolation of individual voices. Preliminary testing of these microphones demonstrated that individual recordings of singers while in an ensemble setting was possible, with negligible amounts of ambient noise and crosstalk. All individual voices were recorded with a minimum of 35 dB greater signal strength than ambient sound.

Phase Two Procedures Participants signed up for recording sessions in octets, comprising two singers per part on soprano, alto, tenor, and bass. As participants entered the room, they were given the informed consent form and asked to read and sign it. Once consent forms were collected, singers were fitted with their individual microphones. Upon completion of the microphone fittings, participants were placed in a modified semi-circular formation with a minimum of two feet of space between each singer. All singers faced a large projection screen. Figure 3.4 shows the arrangement of singers on four riser levels.

Figure 3.4. Standing arrangement of singer participants. Note: Lines represent built-in room risers.

This formation permitted appropriate self-to-other ratio (Daugherty, 2003; Ternström, 1994, 1999) and reduced cross-talk in the individual microphone signals.

The researcher explained the task to the participants:

Hello. Thank you again for agreeing to participate in this study. Today we are exploring current performance practices of choral singers. We will do a brief sound check to verify signal strength from your microphone, and then we will begin the research task. You will sing My Country Tis of Thee four times, and you will be recorded individually and as a group. No identifying information will be attached to your recording outside of your participant number, and the link between your name and participant number will be broken. Recordings will be kept confidential. Does everyone know the melody to My Country Tis of Thee? (All participants stated that they were very familiar with the melody.) An on- screen conductor will provide instructions on when to start and what to sing. Are there any questions before we begin?

The researcher used an in-room projector to display the videos on a large screen using a MacBook Pro laptop computer. The videos were presented in a counterbalanced order, subject

34 to the restriction that both the gestural condition and vowel condition could not change at the same time. This protocol was followed for all sets of performances with the exception of Order 4 due to proctor error. See Table 3.1 for list of orders.

Table 3.1

Phase 2 Experimental Stimuli by Gesture Height and Vowel.

Order 1st 2nd 3rd 4th

Order 1 Low Di Low Dɔ High Dɔ High Di

Order 2 Low Dɔ Low Di High Di High Dɔ

Order 3 High Di High Dɔ Low Dɔ Low Di

Order 4 High Dɔ High Di Low Dɔ Low Di

Order 5 Low Di Low Dɔ High Dɔ High Di

Order 6 High Di High Dɔ Low Dɔ Low Di

Order 7 Low Dɔ Low Di High Di High Dɔ

Note: Low and High refer to Gestural Conditions; Di and Dɔ refer to sung vowel in International Phonetic Alphabet (IPA) notation.

Participants followed instructions of the video conductor regarding the syllable to be performed, and sang in unison at the appropriate octave per voice part for each of the four videos. Upon completion of the recording task, the microphones were collected, and students completed a short questionnaire. The questionnaire (See Appendix C) served to collect demographic information regarding age, major, and solo and choral singing experience. Two additional free response questions pertained to singer perceptions of the effects of conducting on singing: (a) In a general sense, do you feel that gesture affects singing? If so, please explain, and (b) Did you feel that your singing changed in response to the conductor on the video? If yes, please explain.

Phase Two Analysis I analyzed of the digital recordings with Praat software (Boersma & Weenink, 2013). Analysis of notes was set at a sample rate of 100 Hz (one sample each .01 second). Seven notes were chosen on the basis of duration for analysis in each of the four conditions, (see Figure 3.5). The use of longer notes facilitated both singers’ use of vibrato and analyses of tones. Each note analyzed had minimum durations of at least 1 second.

Figure 3.5. Musical selection. Note: Circled notes were selected for data analysis.

For each pitch, I measured 5 aspects of performance: (1) intonation, as measured by cents deviation from the fundamental pitch according to equal temperament, (2) vibrato rate, as measured by cycles per second, (3) vibrato extent, as measured by calculating the difference score in cents of the average peaks and troughs of the vibrato cycle, (4) vibrato duration, as measured by the percentage of time discernable vibrato cycles were present within the length of a tone, and (5) tone quality, as measured by the spectral center of gravity (SPG). For each variable, the mean of the seven tones was calculated to provide a single value per condition for analysis. To establish reliability, two independent observers analyzed 20% of the tones. Data from the observers were compared by computing correlation coefficients for each category of analysis. Participants’ questionnaire responses were coded to identify keywords and response frequency of those keywords. Two independent observers coded 20% of the questionnaires.

Phase Three: Perceptual Analysis The third phase of this study was designed to provide expert listener evaluations of individual and choral sound in response to the gesture height conditions. Participants (N =11) were recruited from graduate choral conducting students at the university. All participants met the following criteria for inclusion: professional experience directing choral ensembles, with at least two years of graduate study in choral conducting/choral music education. Expert listeners (n = 8 males and n = 3 females) had a mean of 14.9 years choral experience, ranging from 8 to 25 years, with a standard deviation of 7.0 years. Listeners also had a mean of 5.6 years of voice study, ranging from 4 to 12 years, with a standard deviation of 2.8 years. After completing a consent form, these participants were given instructions regarding the listening task.

Audio Recordings for Expert Panel Audio recordings representing minimally and extremely different responses were chosen according to acoustic analysis. Long Term Average Spectrum (LTAS) plots of each of the seven group recordings were generated and indicated tone quality adjustments in response to gesture height. Two group recordings with very similar spectral patterns between gesture height and vowel conditions and two recordings with very different spectral patterns between gesture height and vowel conditions were chosen. Figures 3.6 and 3.7 display LTAS plots of differing responses to gesture height. Figure 3.8 and 3.9 display LTAS plots of similar responses to gesture height.

60 Hz) /

Sound pressure level Sound (dB pressure level -20 0 1250 2500 3750 5000 Frequency (Hz) Figure 3.6. LTAS plot of group 4 singing on [dɔ] in high (red) and low (black) gesture conditions Note: Group 4 recordings were used to explore expert listener perceptions of tone quality with differing spectral energy.

60 Hz) /

Sound pressure level Sound (dB pressure level -20 0 1250 2500 3750 5000 Frequency (Hz)

Figure 3.7. LTAS plot of group 4 singing on [di] in high (red) and low (black) gesture conditions Note: Group 4 recordings were used to explore expert listener perceptions of tone quality with differing spectral energy.

60 Hz) /

Sound pressure level Sound (dB pressure level -20 0 1250 2500 3750 5000 Frequency (Hz) Figure 3.8. LTAS plot of group 3 singing on [dɔ] in high (red) and low (black) gesture conditions Note: Group 3 recordings were used to explore expert listener perceptions of tone quality with similar spectral energy.

60 Hz) /

Sound pressure level Sound (dB pressure level -20 0 1250 2500 3750 5000

Frequency (Hz) Figure 3.9. LTAS plot of group 3 singing on [di] in high (red) and low (black) gesture conditions Note: Group 3 recordings were used to explore expert listener perceptions of tone quality with similar energy.

For the individual recordings, acoustic analysis demonstrated intonation and tone quality (as measured by spectral center of gravity) differences between gesture height and vowel conditions. Recordings of all conditions of two females and two males were selected, with one female’s and one male’s recordings demonstrating similar intonation and tone quality measurements and one female’s and one male’s recordings demonstrating different intonation and tone quality measurements. Altogether, the eight group and 16 individual recordings of one phrase each of My Country Tis of Thee were identified for expert panel analysis. See Table 3.2 for a list of the individual recordings.

Table 3.2 Phase 3 Experimental Stimuli for Individual Recordings

Individual Gender Condition Spectral Center of Intonation Recordings Gravity (Hz) (cents)

1 Female 1 (G5-1) High Di 429 13

2 Female 1 Low Di 655 -4

3 Female 1 High Dɔ 436 12

4 Female 1 Low Dɔ 715 -11

5 Female 2 (G1-2) High Di 554 42

6 Female 2 Low Di 484 42

7 Female 2 High Dɔ 794 20

8 Female 2 Low Dɔ 765 27

9 Male 1 (G7-3) High Di 488 -8

10 Male 1 Low Di 551 -31

11 Male 1 High Dɔ 519 4

12 Male 1 Low Dɔ 516 -27

13 Male 2 (G2-4) High Di 349 -68

14 Male 2 Low Di 291 -66

15 Male 2 High Dɔ 504 -52

16 Male 2 Low Dɔ 553 -48

Note: Numbers in parentheses refer to group and track number of participant. Female and Male 1 demonstrated similar performances, while Female and Male 2 demonstrated different performance.

Phase Three Procedures Recordings were presented in a single order, counterbalanced between similar/different pairs, for the listener participants to evaluate. See Table 3.3 for orders. For each excerpt, participants used 1-7 rating scales to indicate their perceptions of intonation, tone quality, vocal freedom, blend, vibrato, and dynamics. Anchor words identified 1 as poor and 7 as excellent (See Figure 3.10). Appendix D shows the full expert-listener rating sheet. The research questions for the present study concerned only intonation, tone quality, vocal freedom, and vibrato: the variables blend and dynamics were included as distractors.

Table 3.3 Order of Expert Listener Recordings

Recording Participant Condition

1. Female 1 (Similar) High Di 2. Female 1 Low Di 3. Male 2 (Different) Low Dɔ Male 2 High Dɔ 4. Group 3(Similar) High Di 5. Group 3 Low Di 6. Male 2 Low Di 7. Male 2 High Di 8. Group 3 High Dɔ 9. Group 3 Low Dɔ 10. Female 2 (Different) Low Dɔ 11. Female 2 2 High Dɔ 12. Male 1 (Similar) High Dɔ 13. Male 1 Low Dɔ 14. Group 4 (Different) Low Di 15. Group 4 High Di 16. Female 1 (Similar) High Dɔ 17. Female 1 Low Dɔ 18. Female 2 (Different) Low Di 19. Female 2 High Di 20. Group 4 (Different) Low Dɔ 21. Group 4 High Dɔ 22. Male 1 (Similar) High Di 23. Male 1 Low Di 24. Group 4 Low Di * 25. Female 1 High Di * 26. Group 3 High Dɔ * 27. Male 2 Low Dɔ *

Note: * Denotes repeated excerpts

1 2 3 4 5 6 7 Poor Excellent

Figure 3.10. Rating scale for expert panel perceptual analysis.

CHAPTER FOUR RESULTS

Purpose Statement

The purpose of this study was to explore possible effects of gesture height, as defined by the location of the horizontal conducting plane, on acoustical and perceptual measures of individual and group performances. The research questions guiding this investigation were:

(1) How do singers perform in response to gesture height? Do singers change (a) intonation (as measured by fundamental frequency), (b) vibrato rate (cycles per second), extent (average distance in cents between peaks and troughs of the vibratory cycles) and duration (percentage of time a discernible vibratory cycle is present), and (c) tone quality (spectral center of gravity)? (2) What are singer participants’ perceptions on the effect of gesture on singing? A post participation questionnaire queried participant perceptions of gestural effects on singing both in a general sense as well as in this specific research task. (3) How do expert listeners rate individual and ensemble performances in response to higher and lower gestures? An expert panel rated the following categories: intonation, tone quality, vocal freedom, and vibrato.

Research Question One: Singer Performance The first research question was addressed during phase two of this project, in which singer participants performed My Country Tis of Thee in groups of 8. Singers were grouped two per part SATB, while watching the stimulus videos of a conductor performing with high and low gestures. For each gesture height, singers performed the excerpt twice, using the syllables [di] and [dɔ]. Singers were recorded individually with personal, head mounted directional microphones, and the ensemble was recorded using quality acoustic microphones arranged in an “x-y “pattern.

Singer Participant Demographics Singer participants (N = 54) were recruited from existing choral ensembles at Florida State University in Tallahassee, Florida. All participants met the criteria for inclusion in this phase of study: undergraduate music majors taking applied voice lessons at the principal level or

42 above, and at least 18 years of age. The researcher recruited participants from choral ensembles in the College of Music by asking for volunteers. Participants made pre-arranged appointments with the researcher. Although 56 participants signed up for appointments to sing in octets, two singers failed to appear for their appointed time. In order to maintain an equal size of eight singers in each ensemble, a research assistant sang in place of these missing singers, but was not recorded individually. In six instances, singers exhibited symptoms of a cold or some other vocal stress that made analysis questionable, and these singers’ data were excluded from analysis. Additional demographic information may be found in Chapter 3.

Performance Analysis Raw data consisted of 5 measures per condition on seven selected tones from the performed melody. All pitches in the melody with duration longer than a quarter note were selected for analysis. The data from each of the seven tones were averaged for each of the four conditions ((a) High Gesture [di], (b) Low Gesture [di], (c) High Gesture [dɔ], (d) Low Gesture [dɔ]) to provide one value for each measure per singer per condition. An alpha level of .01 was used for all statistical analyses. To establish reliability of the individual measurements, two independent observers analyzed 20% of the tones. Mean frequencies, vibrato rates, extents, durations, and spectral center of gravity were compared by computing correlation coefficients for each category of analysis. Reliability coefficients were satisfactory overall (the mean reliability of the measures combined was r =.83). Coefficients for individual measures were: mean frequency (r =.90), vibrato extent (r =.87), vibrato rate (r = .80), and vibrato duration (r =.62). Reliability for the spectral center of gravity was high, r = .98.

Intonation Intonation scores consisted of cents deviation from the fundamental frequency relative to equal temperament tuned to A-440 Hz. The composite mean value for intonation of the high gestural condition across both vowels was -29.77 cents (SD = 19.8). In the low gestural condition for both vowels, the composite mean value was -33.13 cents (SD =18.4). The composite mean score for the [i] vowel in both gesture height conditions was -22.57 cents (SD = 23.0). The composite mean score for the [ɔ] vowel in both gestural conditions was -40.32 cents (SD =16.1). Table 4.1 displays means and standard deviations of intonation scores for gesture height and vowel.

Table 4. 1 Means and Standard Deviations of Intonation Scores according to Gesture Height and Vowel

Mean Condition Standard Deviation (cents)

High Gesture -29.8 19.8 Low Gesture -33.1 18.4

[di] Vowel -22.6 23.0 [dɔ] Vowel -40.3 16.1

High [di] -18.6 23.1 Low [di] -26.5 23.8 High [dɔ] -40.9 18.0 Low [dɔ] -39.8 15.7

A four-way, repeated measures analysis of variance with two between-subjects factors (gender and gesture order) and two within-subjects factors (gesture height and vowel) was calculated to compare intonation performances. Order effects were found for intonation, F (1, 2 44) = 11.74, p < .01, ηp = .21. Composite mean scores for intonation were as follows: Low Gesture First (M = -45.16, SD = 35.34), and High Gesture First (M = -21.41, SD = 32.50). Singers performed approximately 24 cents more flat when watching the low gesture first. Significant main effects were evident for intonation in response to gesture height, F (1,

2 44) = 10.11, p < .01, ηp = .19. Composite mean scores for intonation by gesture height were: High Gesture (M = -31.77, SD = 24.88) and Low Gesture (M =-35.80, SD =23.56). Significant

2 main effects were also evident in response to vowel, F (1, 44) = 130.34, p < .01, ηp = .75. Composite main scores for intonation were: [di] vowel (M =-23.47, SD = 27.44) and [dɔ] vowel (M = -43.10, SD = 21.69). Singers performed approximately 20 cents more flat on the [dɔ] vowel. No significant differences appeared between genders for intonation F (1, 44) = 1.14, p > .01. A two-way interaction occurred between gesture height and gesture order, F (1, 44) =

2 84.8, p < .01, ηp = .66. Figure 4.1 displays mean intonation scores of high and low gestures by

44 gesture order. In both orders, participants sang less flat in the gesture that was presented second. Singers who saw the high gesture first were overall less flat than singers who saw the low gesture first. Additionally, singers were less flat in the second condition (about 12 cents less) by a greater margin when the high gesture was second. None of the other two-, three-, or four- way interactions between variables were statistically significant.

‐10

‐20

‐30

Cents Deviaon ‐40

‐50

‐60 LowHigh HighLow 1st Gesture ‐51.08 ‐24.29 2nd Gesture ‐39.24 ‐18.53

Figure 4.1. Interaction of gesture height and gesture order in intonation

Vibrato Rate Vibrato rate was measured in cycles per second in tones where a discernible vibrato appeared. The composite mean value for vibrato rate in the high gestural condition was 5.30Hz (SD = .55), compared to the composite value for the low gestural condition of 5.27Hz (SD =.58). The composite mean value for the [i] vowel was 5.28Hz (SD = .60), while the [ɔ] vowel was 5.29Hz (SD = .53). Table 4.2 displays means and standard deviations of vibrato rates for the two dependent variables.

Table 4. 2 Means and Standard Deviations of Vibrato Rates according to Gesture Height and Vowel

Mean Condition Standard Deviation (Hz)

High Gesture 5.30 .55 Low Gesture 5.27 .58

[di] Vowel 5.28 .60 [dɔ] Vowel 5.29 .53

High [di] 5.30 .61 Low [di] 5.26 .52 High [dɔ] 5.29 .60 Low [dɔ] 5.28 .56

A four-way, repeated measures analysis of variance with two between-subjects factors (gender and gesture order) and two within-subjects factors (gesture height and vowel) was calculated to compare vibrato rate performances. No order effects were evident for vibrato rate. Results yielded no significant differences in response to gesture height, F (1, 44) = 1.15, p > .01, vowel, F (1, 44) = .65, p > .01, or gender, F (1, 44) = .18, p > .01. No significant interactions occurred in measures of vibrato rate.

Vibrato Extent Vibrato extent was calculated by finding the difference in cents between the mean values of the peaks and troughs during vibrato cycles. The composite mean value for vibrato extent in the high gestural condition was 97.82 cents (SD = 39.22), and in the low condition was 96.71 cents (SD = 40.96). The composite mean value for the [i] vowel was 96.93 cents (SD = 42.07), and the mean value for the [ɔ] vowel was 97.59 cents (SD = 38.53). Table 4.3 displays means and standard deviations of vibrato extent for gesture height and vowel.

Table 4. 3 Means and Standard Deviations of Vibrato Extent according to Gesture Height and Vowel

Mean Condition Standard Deviation (cents)

High Gesture 97.82 39.22 Low Gesture 96.71 40.96

[di] Vowel 96.93 42.07 [dɔ] Vowel 97.59 38.53

High [di] 97.28 42.83 Low [di] 96.58 42.13 High [dɔ] 98.35 37.42 Low [dɔ] 96.84 40.68

A four-way, repeated measures analysis of variance with two between-subjects factors (gender and gesture order) and two within-subjects factors (gesture height and vowel) was calculated to compare vibrato extent performances. Comparisons yielded no order effects for vibrato extent. No significant differences appeared between vibrato extents in response to gesture height, F (1, 44) = .38, p > .01, vowel, F (1, 44) = .12, p > .01, or gender, F (1, 44) = 4.84, p > .01. Significant two-way interactions occurred between gesture height and gesture order, F

2 2 (1, 44) = 8.94, p < .01, ηp = .17, and vowel and gender, F (1, 44) = 10.87, p <.01, ηp = .20. Figure 4.2 displays mean vibrato extent scores of high and low gestures by gesture order. Vibrato extents decreased in the second gesture in both orders, but vibrato extents diminished by a greater amount in the High-Low order. Figure 4.3 displays the mean scores of vowel by gender. Female extents were larger on the [dɔ] vowel, while male extents were larger on the [di] vowel. Additionally, males and females exhibited greater differences between performances of the [dɔ] vowel than in performances of the [di] vowel.

100 95 90 85 80 75

Vibrato Extent in Cents 70 LowHigh HighLow 1st Gesture 98.62 97.24 2nd Gesture 95.9 93.1

Figure 4.2. Interaction of gesture height and gesture order in vibrato extent.

115

110

105

100

85 Vibrato Extent in Cents 80

70 Males Females [di] 85.67 106.2 [dɔ] 80.76 112.24

Figure 4.3. Interaction of gender and vowel in vibrato extent

Vibrato Duration Vibrato duration was calculated by finding the percentage of time a vibrato cycle was discernible within the duration of a selected note. The composite mean value for vibrato duration in the high gestural condition was 80.4% (SD =20.8). The composite mean value for duration in the low condition was 78.9% (SD = 22.2). The composite mean value for the [i] vowel was 80.2% (SD =20.8), and the [ɔ] vowel was 79.1% (SD = 22.9). Table 4.4 displays means and standard deviations of vibrato duration for the two dependent variables.

Table 4. 4 Means and Standard Deviations of Vibrato Duration according to Gesture Height and Vowel

Mean Condition Standard Deviation (Percentage of Note Duration)

High Gesture 80.4 20.8 Low Gesture 78.9 22.2

[di] Vowel 80.2 20.8 [dɔ] Vowel 79.1 22.9

High [di] 80.6 21 Low [di] 79.9 22 High [dɔ] 80.2 24 Low [dɔ] 78.0 24

A four-way, repeated measures analysis of variance with two between-subjects factors (gender and gesture order) and two within-subjects factors (gesture height and vowel) was calculated to compare vibrato duration performances. No order effects were evident for vibrato duration. Comparisons yielded no significant differences in vibrato duration in response to gesture height, F (1, 44) = 1.18, p > .01, vowel, F (1, 44) = .58, p > .01, or gender, F (1, 44) = 5.72, p > .01. No significant interactions were seen in any of the measures of vibrato duration.

Tone Quality For this study, the spectral center of gravity (SPG, also know as spectral centroid) served as the empirical representation of tone quality. The composite mean value for spectral center of gravity in the high gestural condition was 568.4 (Hz) (SD =79.49). The composite mean value for SPG in the low condition was 563.8 (SD = 70.96). The composite mean value for the [i] vowel was 447.8 (SD =101.18), and the [ɔ] vowel was 684.4 (SD = 65.14). Table 4.5 displays means and standard deviations of SPG for the two dependent variables.

Table 4.5 Means and Standard Deviations of Spectral Center of Gravity according to Gesture Height and Vowel

Condition Mean Standard Deviation

(Hz)

High Gesture 568.4 79.5 Low Gesture 564.8 71.0

[di] Vowel 447.8 101.2 [dɔ] Vowel 684.4 65.1

High [di] 453.8 121.1 Low [di] 441.8 93.4 High [dɔ] 683.1 86.8 Low [dɔ] 685.8 63.1

A two-way repeated measures analysis of variance with one between-subjects factor (gesture order) and one within-subjects variable (gesture height) was calculated to compare tone quality performances. Because of inherent differences between vowels and genders in spectral center of gravity (SPG), these comparisons would provide little data of value for this study. No order effects were found for SPG. No significant main effects or interactions appeared in response to gesture height for SPG, All F values were ≤ .53, p > .01.

Research Question Two: Singer Perceptions Following the singing task in phase two, participants completed a short questionnaire (See Appendix C). The questionnaire served to explore two aspects of singer perceptions of gesture: (a) the effect of gesture on singing in a general manner, and (b) the effect of the conductor’s gesture in this specific research task. The two questions required either a ‘yes’ or ‘no’ response, followed by blank lines to provide an optional explanation of their response. Free response answers were coded according to keywords and categorized into emerging themes. An independent observer reviewed 20% of the comments and sorted them into categories. These category assignments concurred with the researcher in 83% of the cases (using the method of agreements divided by number of agreements plus disagreements). The first question, “In a general sense, do you feel that gesture affects singing?” was answered affirmatively by 100% of participants. Regarding musical elements in the free response portion of the questionnaire, participants mentioned dynamics most often (mentioned by 22 participants, 46%), followed by articulation (n = 16, 33%), breath (n = 13, 27%), tone quality (n = 10, 21%), and phrasing (n = 9, 19%). Other keywords mentioned were tempo, vowels, emotion, intonation, consonants, expressivity, facial expression, soft palate height, vocal freedom, inflection/text stress, attitude, ensemble unity, posture, and styles. Table 4.4 displays categorized response frequencies. Regarding non-musical aspects of gesture, 3 participants remarked on the need for some degree of learning for singers to comprehend conductor gestures. The second question, “Did you feel that your singing changed in response to the conductor on the vide0?” was answered ‘yes’ by all but one respondent (98%). Dynamics were again the most frequently listed musical element (17 participants, 35%), followed by tone quality (n = 13, 27%), breath (n = 11, 23%), tempo (n = 6, 13%) and intonation (n = 5, 10%). The remaining categories of responses include articulation, entrances/cutoffs, phrasing, eye contact, energy, freedom, mood, soft palate, technique, tension, and text stress. Table 4.6 displays categorized response frequencies. Verbatim questionnaire responses can be found in Appendix E and Appendix F.

Table 4.6 Frequency of Musical Elements Listed in Response to Participant Questionnaire

1) In a general sense, do you feel that gesture 2) Do you feel that your singing changed in affects singing? response to the conductor on the video?

22 Dynamics 17 Dynamics 16 Articulation 13 Tone Quality 13 Breath 11 Breath 10 Tone Quality 6 Tempo 9 Phrasing 5 Intonation 6 Tempo 4 Articulation 6 Vowel 4 Entrances/Cutoffs 5 Emotion 4 Phrasing 5 Intonation 3 Eye Contact 3 Consonants 1 Energy 3 Expressivity 1 Vocal Freedom 3 Facial Expression 1 Mood 3 Soft Palate Height 1 Soft Palate Height 2 Vocal Freedom 1 Technique 2 Inflection 1 Tension 1 Attitude 1 Text Stress 1 Ensemble Unity 1 Posture 1 Style

Research Question Three: Expert Listener Perceptions Phase three of this study was designed to investigate perceptions of an expert panel on ensemble and individual singer performances in response to gesture height. Raw data consisted of expert listener evaluations using a rating scale ranging from 1 to 7 in five categories: (a) intonation, (b) tone quality, (c) vocal freedom, (d) blend, (e), vibrato, and (f) dynamics. The scale was anchored with the terms poor and excellent. The research questions for the present

52 study concerned only intonation, tone quality, vocal freedom, and vibrato. The variables blend and dynamics were included as distractors.

Expert Listener Reliability In repeated observations of the same group and individual excerpts, the 11 judges exhibited a somewhat low intra-observer reliability, r =.64. Inter-judge reliability was calculated with Kendall’s Coefficient of Concordance for each measure in both group and individual performances. Judges exhibited low agreement in ratings of group performances, Kendall’s W = .22, and were somewhat higher in their ratings of individual performances, Kendall’s W = .53.

Analysis of Ensemble Performance Four ensemble recordings were chosen on the basis of LTAS plots to represent similar and different tone qualities in response to low and high gestures (LTAS plots may be found in Chapter 3). Table 4.7 displays means and standard deviations for the ratings of group recordings by the panel of expert listeners. When comparing pairs of the similar LTAS group, recordings of the [di] vowel were rated higher in all categories for the high gesture condition than in the low gesture condition. For the [dɔ] vowel, ratings were similar between conditions. For the different LTAS examples (see Table 4.8), ratings were close between gesture height conditions in both vowels.

Table 4.7 Means and Standard Deviations of Expert Listener Ratings of Similar Group Performances

Condition Intonation Tone Quality Vocal Vibrato Freedom

M SD M SD M SD M SD

High [di] 4.8 1.3 4.0 1.15 4.4 1.4 3.8 1.0

Low [di] 3.8 1.5 3.5 1.2 3.7 1.4 3.7 1.2

High [dɔ] 4.5 1.8 4.5 1.6 4.3 1.7 4.2 1.5

Low [dɔ] 4.7 1.4 4.5 1.2 4.5 1.1 4.2 1.4

Table 4.8 Means and Standard Deviations of Expert Listener Ratings of Different Group Performances

Condition Intonation Tone Quality Vocal Freedom Vibrato

M SD M SD M SD M SD

High [di] 4.8 1.5 4.7 1.0 4.6 1.2 4.2 1.3

Low [di] 5.0 1.6 4.8 1.1 4.2 1.1 4.4 1.3

High [dɔ] 4.9 1.3 4.8 1.2 4.8 1.0 4.5 1.1

Low [dɔ] 4.9 1.3 5.2 1.3 4.7 1.0 4.5 0.9

Statistical analysis confirmed that ratings were very close between conditions. A chi- square analysis showed that listeners did not discriminate tone quality between the low gesture and high gesture group performances of the [di] vowel, Χ2 (1, 11) = .02, p > .5. Similarly, there were no differences between ratings of gesture conditions in the [dɔ] vowel, Χ2, (1, 11) = .04, p > .5. For ratings of intonation, analysis again showed no significant difference between gesture condition for the [di] vowel, Χ2 (1, 11) = 1.03, p > .3, nor for the [dɔ] vowel, Χ2 (1, 11) = .01, p > .9. No significant differences were evident in ratings of vocal freedom in either condition in either vowel: [di] vowel, Χ2 (1, 11) = .01, p > .9; [dɔ] vowel, Χ2 (1, 11) = .01, p > .9. For ratings of vibrato, listeners did not differentiate between gesture conditions in either vowel: [di] vowel, Χ2 (1, 11) = .01, p > .9; [dɔ] vowel, Χ2 (1, 11) = .03, p > .8.

Analysis of Individual Performance In addition to the ensemble recordings, one female’s and one male’s individual recordings were chosen to represent singers whose intonation and tone quality data demonstrated very similar responses to conducting gestures. Conversely, one female’s and one male’s individual recordings demonstrated differing responses to conducting gestures. Four recordings per singer (representing both gestural and vowel conditions) were used for the expert panel to evaluate.

Tables 4.9 and 4.10 display means and standard deviations of expert ratings of the similar and different female performances, respectively. Tables 4.11 and 4.12 display means and standard deviations of expert ratings of the respective similar and different male performances. No distinct trends emerge when looking at vowels or gesture height in either of the female performances. In both similar and different male performances, the low gesture condition of the [dɔ] vowel was rated higher across all measures. The Wilcoxon matched-pairs test showed that this pattern was statistically significant, T = 36, p = .01, z = 2.53.

Table 4.9 Means and Standard Deviations of Expert Listener Ratings of Similar Female Performances

Condition Intonation Tone Quality Vocal Freedom Vibrato

M SD M SD M SD M SD

High [di] 3.5 1.0 4.3 1.1 3.5 1.6 3.7 1.3

Low [di] 3.8 1.1 3.6 1.3 3.2 1.25 3.5 1.0

High [dɔ] 4.7 1.6 4.3 1.2 3.7 1.3 3.7 1.3

Low [dɔ] 4.6 1.5 4.3 1.1 4.0 1.3 3.8 1.2

Table 4.10 Means and Standard Deviations of Expert Listener Ratings of Different Female Performances

Condition Intonation Tone Quality Vocal Freedom Vibrato

M SD M SD M SD M SD

High [di] 4.8 1.4 4.7 1.3 4.9 1.2 4.8 1.8

Low [di] 4.5 1.4 4.4 1.6 4.9 1.4 4.5 1.8

High [dɔ] 5.0 0.9 5.5 1.1 5.1 0.9 5.3 1.2

Low [dɔ] 3.1 1.3 5.4 1.1 5.4 1.1 5.5 1.1

Table 4.11 Means and Standard Deviations of Expert Listener Ratings of Similar Male Performances

Condition Intonation Tone Quality Vocal Freedom Vibrato

M SD M SD M SD M SD

High [di] 3.5 1.0 3.6 1.3 3.4 0.9 2.8 1.2

Low [di] 3.5 1.0 3.4 1.0 3.0 0.9 2.6 1.2

High [dɔ] 3.1 0.9 3.3 1.0 3.1 1.5 3.1 1.2

Low [dɔ] 3.7 1.3 3.5 1.2 3.4 1.5 3.2 1.0

Table 4.12 Means and Standard Deviations of Expert Listener Ratings of Different Male Performances

Condition Intonation Tone Quality Vocal Freedom Vibrato

M SD M SD M SD M SD

High [di] 2.7 1.3 2.4 1.2 2.5 1.4 2.7 1.5

Low [di] 2.7 1.3 2.6 0.9 2.5 1.4 2.9 1.4

High [dɔ] 2.2 1.5 2.6 1.2 2.9 1.4 2.5 1.4

Low [dɔ] 2.6 1.5 3.1 1.1 3.1 1.2 3.0 1.3

Statistical analysis of the individual ratings yielded very similar findings to the expert ratings of the group recordings. A Χ2 analysis showed that listeners did not discriminate tone quality between the low gesture and high gesture individual performances of the [di] vowel, Χ2 (1, 11) = .01, p > .9. Likewise, there were no differences between gesture conditions of the [dɔ]

56 vowel, Χ2 (1, 11) = .02, p > .8. For intonation ratings, no significant differences appeared between gesture conditions in the [di] vowel, Χ2 (1, 11) = .01, p > .9 or the [dɔ] vowel, Χ2 (1, 11) = .05, p > .8. Neither did listeners differentiate performances in ratings of vocal freedom: [di] vowel, Χ2 (1, 11) = .01, p > .9, [dɔ] vowel, Χ2 (1, 11) = .04, p > .8. Again, no differences emerged between gesture conditions in ratings of vibrato: [di] vowel, Χ2 (1, 11) = .01, p > .9, [dɔ] vowel, Χ2 (1, 11) = .06, p > .8.

CHAPTER FIVE

DISCUSSION

Purpose Statement

The purpose of this study was to explore the effects of gesture height, or the location of the horizontal conducting plane, on acoustical measures of individual and group performances in response to gesture, as well as expert listener perceptions of performed intonation, vibrato, tone quality, vocal tension, and dynamics. The following questions guided this investigation: 1) How do singers perform in response to gesture? Do singers change: (a) intonation (as measured by fundamental frequency) (b) vibrato rate (cycles per second), extent (average cents distance between peaks and troughs of the vibratory cyles) and duration (percentage of time a discernible vibratory cycle is present) (c) tone quality (spectral center of gravity)? 2) What are singer participants’ perceptions of the effect of gesture on singing? 3) How do expert listeners rate individual and ensemble performances in response to higher and lower gestures in the following categories: intonation, vibrato, tone quality, vocal tension, and dynamics?

Summary of Results Question 1. Singer performance in response to gesture height

Singers evidenced no changes in measures of vibrato rate, vibrato duration, or tone quality (SPG). In vibrato extent, an order effect was seen whereby singers reduced their vibrato extents in the second gesture, but by a higher magnitude in going from the high to low gesture, compared to the sequence of low to high. Singers exhibited the greatest change in measures of intonation; participants tended to perform less flat in the high condition than in the low. Order effects were also seen in intonation: singers performed less flat while performing with the second gestural condition, no matter whether it was high or low.

Question 2. Singer perceptions of the effect of gesture

Singers unanimously agreed that conducting gestures have an effect on singing in the general sense, most frequently citing dynamics, articulation, breath, and tone quality as specific factors.

Singers again agreed (98% positive response) that gesture affected their singing in the specific task for this study. Response frequencies were highest for dynamics, tone quality, breath, and tempo.

Question 3. Expert listener perceptions

The listening panel did not discriminate between gesture conditions in the group recordings, or in the female individual recordings. Listeners did not discriminate between gesture conditions of the male recordings for the [di] vowel. In the male recordings of the [dɔ] vowel, expert listeners rated all rated measures of the low gesture higher than the high gesture.

General Discussion The central question in this study considers whether gesture change might elicit a differential response without verbal explanation. Does a gesture prompt a common reaction from singers, or must it be accompanied by a verbal definition explaining its purpose? Much of the gestural language employed by conductors has implied meaning, whether explained explicitly by conductors, i.e. “This gesture means get soft!” or learned through reinforcement behaviors from conductors. For example, a conductor might not specifically verbalize a desire for non-vibrato singing. However, imagine that after the conductor employs a gesture mimicking a smoothing motion on a tabletop, singers then use less vibrato. A positive affirmation from the conductor following this behavior provides a learned association for that gesture. The smoothing gesture now implies less vibrato for those choristers. This study aimed to ascertain whether there might be performance associations connected to height of gesture. Future research might explore whether such possible associations are learned or implicit.

Question 1. Singer Performance in Response to Gesture Height Do singers change intonation? Participants performed less flat in the high gesture condition than in the low gesture condition, but only by approximately 4 cents when considering overall means (High Gesture M = -31.77; Low Gesture M = -35.80). While the difference was statistically significant, it might not be musically meaningful. Perhaps the more striking result concerns the order effect seen in intonation. Singers who saw the high gesture first performed markedly less flat in both conditions than singers who saw the low gesture first (24 cents flat in the high condition compared to 51 cents flat in the low condition). In both orders (High-Low and Low-High), participants sang less flat in the second condition. This might be attributed to a novelty effect.

Singers noticed a change in the gestural language, which may have served as a cue to “perk up.” Future study might explore the novelty effect of gestural language using conducting emblems. These intonation results do not concur with the recent gesture studies by Brunkan (2013) and Grady (2013) where low gestures elicited more accurate intonation. The difference here may lie in the location of the high gesture. In this study, the position of the high gesture was between the clavicle and the chin. In the Brunkan and Grady studies, the high gesture was located above the chin. Future study might incorporate a third location to further explore these gestural effects on intonation. Another possible explanation for the differences in intonation results could be the type of gesture used in Brunkan’s study. The gesture producing the best intonation was an upward moving circular motion. The traditional pattern used in this study incorporates a downward moving ictus, which might have a possible effect on intonation performance. This study also explored gestural effects in two different vowel conditions. Though no interaction appeared between vowel condition and gesture height, singers did perform the two vowel conditions with significantly different pitch levels. The [di] vowel (M = -23.47) was performed considerably less flat overall than the [dɔ] vowel (M =-43.10). Such differences in vowel performances warrant further exploration of the effect of gesture on many vowels. Comparison of two forward vowels or two back vowels might provide more information regarding these results.

Do singers change vibrato performance? While no main effects appeared in response to gesture height or vowel for any of the vibrato measures, two interactions occurred in measures of vibrato extent. Singers reduced their vibrato extents in the second gesture condition by 2-4 cents. Although this change evidences statistical significance, listeners would not hear any distinction between the two, and the difference is therefore not meaningful to listener perception. However, the order effect seen here further substantiates the possibility of a novelty effect. Once more, it appears that singers respond to a change in gesture rather than a specific gesture. The second interaction seen in vibrato extent, between vowel and gender, might provide useful data for another study, but is not germane to the research questions of this study. Mann’s previous study (2014) concerning gesture and vibrato demonstrated that singers used a smaller vibrato extent while watching a primarily horizontal pattern (a traditional 4/4 pattern with very small rebound) compared with a primarily vertical pattern (a pattern incorporating very large rebound). In that study, singers who saw the horizontal pattern first performed with a mean vibrato extent of 83 cents. In contrast, singers who saw the vertical

60 pattern first performed with a vibrato extent of 110 cents. This marked difference in vibrato performances indicated a notable effect of gesture on vibrato, though there were some alternative possible explanations with respect to order and design. The present study did not demonstrate such differences, indicating that size of rebound might hold more influence on vibrato extent than gesture height. Future study might compare the two variables of conducting plane and rebound height directly, to explore this premise further. Because of the somewhat controversial nature of vibrato in choral music, knowledge of the effects of gesture in this arena would be of benefit to choral directors and vocal pedagogues.

Do singers change tone quality (spectral center of gravity)? A working hypothesis for this study operated on the pedagogical assertions that conducting near the abdomen elicits better breath efficiency from singers, and that conducting near the larynx elicits laryngeal tension or a raised larynx. These hypotheses predicate a change in tone quality. The fact that spectral center of gravity measures between gesture heights evidenced no differences does not support the pedagogical literature regarding the influence of gesture on tone. It may be that spectral center of gravity does not reflect the refined processes of the human brain in perception of tone quality, or that the limitations of this study may have obscured finding the results described anecdotally by conducting pedagogues. Logically, of course, it could also be that gesture height has no replicable effect on tone, or at least only in particular contexts. However, this line of inquiry ought to be revisited, perhaps with additional methods of measurement. Further, Brunkan (2012) stated that exposure to a gesture over time may be an important factor in gestural impact on singer performance. One singer participant in this study remarked, “I sang louder as I got more accustomed to the conductor.” The idea of exposure over time was taken into consideration when designing the present study. Singers performed the excerpt for two verses in each condition with analysis beginning on the second verse, in order to give time for the gestural condition to begin to have an effect. However, one verse of a gesture may not be sufficient time for gestural effects to take place. These results differ from the Daugherty and Brunkan (2012) study and the Manternach (2012) lip rounding study where singers evidenced change in four measures of music. Such immediate effects were not seen in the present study. Future study on gesture height might consider longer exposure to a gesture before analysis takes place.

Question 2. Singer Perceptions of the Effect of Gesture Participants in this study, in response to questions regarding both the effect of gesture on singing in a general sense and the effect of gesture on singing in this study, listed dynamics most frequently as a specific factor influenced by gesture. After that, the answers diverged quite a bit, which is not unexpected, as the conductor purposely displayed very little musical intent in these videos outside of entrances and cutoffs. Researchers have examined the efficacy of intended communications by conductors, and have found that conductors have varying measures of success based on their ability to encode nonverbal messages (Byo, 1990). Studying the conductor’s ability to communicate musical intent was not the goal of this project, so the diversity of responses to questions about general conducting and questions about this specific task has little bearing on the implications of this study. This study was concerned with implicit effects resulting from gesture height, as the conductor never defined the intent of his gestures. The high participant response frequency of tone quality and breath confirm that singer perceptions concur with the pedagogical literature asserting the influence of gesture on these variables. Singers do appear to perceive that gesture affects performance. Some singers found gesture height to function as a reminder of various aspects of vocal technique. One singer remarked, “When the gesture was higher, I felt tension in my larynx. I could feel my tongue raise. When the gesture was lower, I felt that my breath could be freer to move, and I was overall more relaxed.” Another singer remarked, “The second two times, the conductor’s hands were higher and his shoulders were more tense. I leaned toward a brighter, more forward tone just naturally, but also tended to forget my breath.” Contrarily, another singer remarked that the higher gesture served as a cue to lift the soft palate, indicating that in this instance the high gesture functioned in a positive way for the singer’s technique. Another participant found beneficial attributes of both conditions, stating, “While watching lower conducting, I felt more apt to sing technically correct, however, higher conducting seemed to influence more lifted pitch to my ear.” These comments serve as a reminder that what cannot be ignored in a study about the implicit nature of a gesture is the history of learned associations brought to the task by the singer participants. Future study with beginning singers might further explore possible influences of implicit and learned associations. An unanticipated effect of gesture height was brought to light by several remarks from singer participants on post-participation questionnaires. Some singers experienced greater “connection” to the conductor when the gesture height was higher. A few remarked specifically that when the gesture was higher, they found it easier to see the conductor’s face, and could therefore more easily interpret the conductor’s musical intentions. One singer stated, “It’s

62 harder to sing musically when his hands were low. I want to see both his hands and his expression.” Another participant remarked,

When his hands were up closer to his face I was able to get his mood and feel for the piece way better than when they were lower. I had to really try for tempo more when his hands were lower; it was harder to keep in time. When it was up by his face it was a lot clearer for many things like where he wanted us to breathe and marking like ritards and word stress/phrasing. Plus his face helped me understand what he wanted overall for the piece.

Others appeared not to notice the proximity of the hands to the face, but instead remarked that the conductor’s affect was robotic in the low gesture condition and encouraging in the high gesture condition, despite a priori expert agreement that the conductor’s facial affect was consistent between conditions. This concurs with Nowak and Nowak’s (2002) assertion that the conducting plane should be high enough to coordinate with eye contact. Future study concerning proximity of gesture to facial expressions is warranted, especially in regard to expressive intent.

Question 3. Expert Listener Perceptions Expert listener ratings of group performances did not evidence statistical differences between conditions. The trend of these scores demonstrates that listener perceptions did not align with the LTAS analysis of the group recordings. The group with similar LTAS plots between conditions elicited slightly different ratings between conditions, while the group with differing LTAS plots between conditions elicited very similar ratings between conditions. In the individual ratings, a trend emerged among the male singers performing the [ɔ] vowel, whereby their performance ratings in the low gesture condition were consistently higher than their ratings in the high gesture condition. This trend yielded statistical significance for the [dɔ] vowel. The [ɔ] vowel is notoriously more difficult for young singers than the [i] vowel, and in this study was performed significantly more flat than the [i] vowel. Because of the difficult nature of the [ɔ] vowel in vocal technique, any improvement according to listener perceptions would appear to be valuable information to choral directors and pedagogues. Across the scope of this study, expert listeners heard a change between gesture heights in only the individual male recordings selected. No change was evident between the individual female singers or the two sets of group recordings. Expert perceptions may have yielded

63 different results if listeners were asked to make a choice between matched pairs; a paired- comparison design might have detected a subtle preference for low or high gestures that perhaps was lost in the individual rating scales. The low interjudge and intrajudge reliability of the expert panel corroborates the findings of Wapnick and Ekholm (1997) that due to its subjective nature, tone quality is difficult to quantify. Judges in this study were not given a rubric to define any of the measurement categories. In future studies, a rubric might contribute to stronger inter- and intra-judge reliabilities.

Recommendations for Future Research Younger singers, or older singers with less training, might be more susceptible to possible effects of gesture height than more experienced singers. One singer participant remarked, “I feel like since I already have an idea of what healthy singing feels like, I won’t compromise that because someone’s gesture is too high. It (their high gesture) doesn’t help certainly, but it might affect young/less experienced singers more.” Future study might consider a replication of the present study with high school or younger singers, or with community singers with little vocal training, as well as singers with moderate and extensive amounts of vocal and choral experience. For example, is there a point at which singers ignore gestural commands in favor of their own ideas of vocal technique or singer health? Or, as singers develop greater habit strength in their vocal technique, do gestures that have been shown to elicit tension have less effect? Research exploring the idea of mimicry and empathetic response has found that singer body movement corresponds highly with conductor body movement (Manternach, 2011), and that conductor gestures can elicit unwanted tension in singers (Fuelberth, 2003b, 2004). As stated earlier, one of the primary research questions of the present study concerned the proximity of the gesture to the larynx: it was reasoned that a high gesture (near the larynx) would elicit laryngeal tension, in much the same manner as the fisting and stabbing gestures were shown to evoke perceptions of tension among singers in Fuelberth’s studies. Future study might focus specifically on the gesture concentrated near the larynx, with the conductor exhibiting varying amounts of tension in the neck and shoulders. General discussion of the conducting plane takes place in terms relative to the conductor’s body. In the present study, clavicle to chin defined the high condition, and navel to bottom of rib cage defined the low condition. Lamb’s (1974) text suggests that arms be raised high enough for visibility by the ensemble and low enough to be comfortable. In a very tall

64 person, or a conductor with a long torso, the possibilities for location of the conducting gesture are greater than in a person with a shorter torso. Future study might examine effects of gesture height in persons of varying heights. This study also concerned the idea that gestures might be learned or implicit. Sousa (1988), Mayne (1992) and Cofer (1994, 1998) demonstrated that many conducting gestures establish emblem status, with the number of emblems performers recognize increasing with experience. These emblem studies examined gestures indicating specific musical intent, such as articulation or dynamics. There exists a paucity of research, however, regarding the effect of gesture on less defined variables, such as breath management or tone quality. Napoles (in press) demonstrated that verbal instructions elicited stronger reactions than gestures in word stress and articulation. Future study might expand Napoles’ study to explore breath management and tone quality. In studies on rehearsal time use, the 1-2-3 teaching cycle (1 – teacher direction, 2 - student response, 3 - teacher feedback) has been shown to be an efficient and effective means of instruction (Yarbrough & Madsen, 1998). Future study might compare non-verbal teacher directions paired with teacher’s specific, related feedback to explore learned and implicit associations with specific conducting gestures in beginning singers.

Conclusion This study does not offer substantial evidence in support of pedagogical literature regarding gesture height. In the acoustical analyses, presentation order interacted with intonation and vibrato; differences were shown in response to the second gesture condition. What the order effects suggest, however, is that a change in gesture may well elicit a novelty effect. Singers appear to respond to a change in gesture rather than a specific gesture. Such information may be of use to conductors in informing their conducting practices in rehearsal and particularly in performance. Singer perceptions of gesture provide useful information; perhaps the most important aspect of nonverbal communication is whether gestures function as intended. Participants indicated a variety of perceptions of gesture height in this study. Expert listeners’ preference for the [ɔ] vowel in the low gesture height in the individual male recordings is consistent with pedagogical claims of the benefits of low gestures (connection to the breath, freedom in the larynx). Many characteristics of the effects of gesture height remain to be investigated.

APPENDIX A

HUMAN SUBJECTS COMMITTEE APPROVAL MEMORANDUM

Office of the Vice President for Research Human Subjects Committee Tallahassee, Florida 32306-2742 (850) 644-8673 · FAX (850) 644-4392

APPROVAL MEMORANDUM

Date: 01/09/2014

To: Lesley Mann

Address: 3433 Castlebar Circle, Tallahassee, FL 32309

Dept.: MUSIC SCHOOL

From: Thomas L. Jacobson, Chair

Re: Use of Human Subjects in Research THE EFFECT OF CONDUCTOR’S GESTURE HEIGHT ON ACOUSTICAL AND PERCEPTUAL ANALYSIS OF INDIVIDUAL AND CONGLOMERATE CHORAL SOUND

The application that you submitted to this office in regard to the use of human subjects in the proposal referenced above have been reviewed by the Secretary, the Chair, and two members of the Human Subjects Committee. Your project is determined to be Expedited per 45 CFR § 46.110(7) and has been approved by an expedited review process.

The Human Subjects Committee has not evaluated your proposal for scientific merit, except to weigh the risk to the human participants and the aspects of the proposal related to potential risk and benefit. This approval does not replace any departmental or other approvals, which may be required.

If you submitted a proposed consent form with your application, the approved stamped consent form is attached to this approval notice. Only the stamped version of the consent form may be used in recruiting research subjects.

If the project has not been completed by 01/08/2015 you must request a renewal of approval for continuation of the project. As a courtesy, a renewal notice will be sent to you prior to your expiration date; however, it is your responsibility as the Principal Investigator to timely request renewal of your approval from the Committee.

You are advised that any change in protocol for this project must be reviewed and approved by the Committee prior to implementation of the proposed change in the protocol. A protocol change/amendment form is required to be submitted for approval by the Committee. In addition, federal regulations require that the Principal Investigator promptly report, in writing any unanticipated problems or adverse events involving risks to research subjects or others.

By copy of this memorandum, the chairman of your department and/or your major professor is reminded that he/she is responsible for being informed concerning research projects involving human subjects in the department, and should review protocols as often as needed to insure that the project is being conducted in compliance with our institution and with DHHS regulations.

This institution has an Assurance on file with the Office for Human Research Protection. The Assurance Number is IRB00000446.

Cc: John Geringer , Advisor HSC No. 2013.11392

APPENDIX B

APPROVED SINGER PARTICIPANT CONSENT FORM

APPENDIX C

INDIVIDUAL MICROPHONE SPECIFICATIONS

APPENDIX D

SINGER PARTICIPANT QUESTIONNAIRE

Participant)Number______) Group)Number)______) ) Choral'Performance'Study' Singer'Participant'Questionnaire' ) ) Year)in)School)))))))))))) ) Age))))))))))))) ) Major))))))))))))) ) How)many)years)have)you)studied)voice?))))))))) ) How)many)years)have)you)sung)in)choir?)))))))) ) Questionnaire:))

1) In)a)general)sense,)do)you)feel)that)gesture)affects)singing?))Circle)Y)or)N))

2) If)yes,)please)describe)in)detail.)Continue)on)the)back)if)needed.))

______

______)

))))))))))))

3) Did)you)feel)that)your)singing)changed)in)response)to)the)conductor)on)the)

video?)Circle)Y)or)No))

4) If)yes,)please)describe)in)detail.)Continue)on)the)back)if)needed.)))

)))))))))))

))))))))))))

APPENDIX E

EXPERT LISTENER RATING SHEET

Year in School

Circle One:

1. Male Female

2. Music Major Non-Music Major

Musical Area:

Winds Choral Strings Keyboard Other

Degree Program

All Participants:

1. How many years have you participated in choir? ______

2. Have you ever taken voice lessons? Y or N

If yes, how many years?

3. Have you participated in other types of ensembles? Y or N

If yes, please describe:

For each of the following excerpts, please rate the following musical elements on the following scale:

1 2 3 4 5 6 7 Poor Excellent

Practice 1 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7

Practice 2 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 For each of the following excerpts, please rate the following musical elements on the following scale:

1 2 3 4 5 6 7 Poor Excellent

Excerpt 1

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 2

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 3 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 4

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 5 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 74

1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 6

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 7 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 75

1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 8

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 9 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7 76

Dynamics 1 2 3 4 5 6 7 Excerpt 10 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 11

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 12 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 13

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 14 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 15

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 79

1 2 3 4 5 6 7 Excerpt 16 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 17

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 80

Excerpt 18

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 19 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7

Excerpt 20 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 21

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7

Excerpt 22 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 23

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7 83

Dynamics 1 2 3 4 5 6 7

Excerpt 24

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7 Excerpt 25 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7 84

Dynamics 1 2 3 4 5 6 7 Excerpt 26 Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7

Excerpt 27

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7

Excerpt 28

Intonation 1 2 3 4 5 6 7

Tone Quality 1 2 3 4 5 6 7

Vocal Freedom 1 2 3 4 5 6 7

Blend 1 2 3 4 5 6 7

Vibrato 1 2 3 4 5 6 7

Dynamics 1 2 3 4 5 6 7

APPENDIX F

PARTICIPANT VERBATIM COMMENTS (GENERAL)

Open-Ended Response Area Following the Question: ‘In a general sense, do you feel that gesture affects singing? If yes, please describe in detail:’

I feel like since I already have an idea of what healthy singing feels like, I won’t compromise that because someone’s gesture is too high. It (their high gesture) doesn’t help certainly, but it might affect young/less experienced singers more.

Conductors use gesture to show the choir dynamic and intensity levels in the music, any seasoned choir member can read the conductor’s gestures. I think it’s a learned skill, however. I don’t think children of non-singers react the same way.

The facial affect and size of gesture definitely change any approach to choral singing. Obviously, the speed that the director chooses also changes how resonant I can be on certain pitches.

Everything about a gesture affects singing because of the human tendency for coordination. More legato movements inspire a more lyrical tone and vice versa.

The conductor’s gesture is pivotal to achieving the desired sound. It affects breath engagement, articulation, tempo, vowel, and more.

A subtle change in gesture can overtly affect the tone of an ensemble, both positively and negatively. It’s important with gesture not just to beat time and express, but also encourage healthy singing and proper ensemble practices.

For an attentive choir, gesture affects every part of singing. Choir members derive their posture, breathing, attitude, and tone from the conductor. Whether they realize it or not.

Lower ictus provides a deeper breath intake. You feel more connected to the entire body.

While I haven’t studied conducting, I feel as a singer that a conductor’s gestures directly affect my inflection and dynamics.

I believe it affects singing because if a conductor is more expressive, then the singers will respond accordingly, and the same in reverse.

If the conductor conducts on a horizontal plane the singing is more connected. When the conductor pauses that make us stop or lift.

The higher he held his hands and larger the gestures opened up the singers making them sing louder, and the opposite is true when he held his hands lower and gestured smaller. I sang louder as I grew more accustomed to the conductor also.

When he was conducting with a lower gesture, I felt more inclined to sing with a lower soft palette and lower breath. The opposite is true for a higher gesture. A conductor’s gesture can determine whether I breathe in the correct place, or whether I carry through the phrase. It can also determine whether I sing with a specific dynamic or onset.

The affect used by a conductor can act as a parallel to negative/positive reinforcement for a singer. It is as a result of singing with regard to a conductor’s interpretation that singers result to a default form of singing. It is similar to that of a singer who is still depending on sheet music to sing a piece. Because we use the separate hemispheres of the brain to process reading music to sing with a recent change to our technique, things thus occupy our use of the processing part of the brain become convoluted and technique suffers.

Gesture is extremely important in singing, it can be responsible for tone quality, vowel placement, consonant placement, phrasing, tuning, rhythmic integrity, etc. It can also encourage breath support and free singing.

Gestures allow another level of communication between the choir and conductor. They may affect phrasing, dynamics, tempo rubato, or any number of musical ideas.

The hand can affect vowels, breath, facial expression. If the hand is curved it promotes the soft palette to be raised which is god technique; it its flat, the soft palette is lowered and can cause intonation issues, bad vowels, etc. Good gesture promotes good singing.

Yes it can affect the choirs volume and smoothness (legato/staccato).

The lower hand gestures insinuate quieter singing.

Conductor can help shape phrases and encourage light singing verses more forceful, heavy production.

Yes, I believe that it affects singing. Whether it be a legato line or a more rhythmic line. A lowered ictus might affect intonation in the choir. The tension in the arms or face might also transmit to the singers.

The lower ictus allowed shallow breathing and a darker tone quality.

It affects articulation, use of the breath, dynamics, consonant, and so much more. A short rhythmic gesture will create staccato rhythmic singing.

It can guide use of the breath, the tone and vowels and their placement, and great conductors can even influence how healthy you sing.

I feel that gesture affect singing in that certain conducting gestures can make a choir sing differently. If a conductor conducts bigger or smaller, a choir may sound louder or softer. Tension in conducting can cause singers to control their voices more, etc.

Gesture affects when to breath, dynamics, overall expression, and tempo. How big, small, low, or high gesture are indicate how the director wants something to be sung.

Distinct start and stopping points for notes affects overall rhythm. Lower and smaller movements created quieter sound. The gestures affect singing in a positive way.

Yes, usually it affects breathing, dynamics, and the way musical phrases are shaped.

I think that the gesture affects singing because as a choir member that is the only way I know what the conductor wants from the group. It also helps me to stay in sync with the rest of the ensemble.

The gestures of the conductor definitely affects singing. For me, it’s mostly the desired phrase structure/when to breath. It also gave me an indication to sing softer when he put his hands lower.

Gesture controls about any variable in singing; it affects the performance in things like tempo and dynamics, controls the levels of articulations, but also significantly affects the timbre of a choir, and brings out expression.

The first recording had higher hand gestures and caused me to sing louder. The second had lower gestures so I sand quieter.

If a conductor gestures toward, he wants to build, if he gestures back he wants you to move less, be quiet. It make the choir sound good because it reminds you to move and go somewhere rather than staying stagnant, just singing notes, etc.

When one sees the conductor’s hand give a stronger, more pointed gesture, one sings with stronger expression. Though if one sees a more fluid gesture, it can indicate more legato.

Depending on the conductor’s gesture, the choir will sing loud, soft, light, heavy, rhythmic, legato, restrained, free, etc. For example, a high ictus indicates a softer dynamic level…though it there is no variation in the ictus height then it will lose overall effectiveness.

I believe gesture affects singing because if the conductor feels you need more emotion then he/she will gesture it, and make the choir sing that emotion.

The way a conductor lifts his arms to cutoff the choir of the height of his arms when conducting. High ictus = soft dynamic Low ictus = loud dynamic

Assuming the singer is trained to follow the conductor, gesture absolutely affects singing. I interpret the slight movements of the hand, arm, or face into however I feel the music should be delivered.

The gestures of the conductor tell the choir how he/she wants them to sing the piece. They give the choir a clear understanding of the style. Dynamics, articulation, etc.

The more subtle the gesture, then the more subtle/quiet I will sing. And vice versa for big gestures/ strong and loud singing.

I respond very highly to conductor’s gestures. If I practice, gestures serve a s a visual reference of what happens inside. But my gestures would be different in a solo performance.

Gestures to conducting are like someone speaking to you. They must conduct in a meaningful way. If someone said “I need to go to the hospital” in a monotone voice, you might not believe them. But if they scream “I NEED TO GO TO THE HOSPITAL” it is different!

High/low ictus eights = dynamic contrast and can affect breath control. Gesture also affects articulation (rhythmic, legato, etc.) phrasing, vowel formation, and diction (aggressive vs. passive) depending on what the hands and body are showing.

There is a lot that gesture has to do with singing. It can affect tempo, dynamics (by how big the gesture is), word stress, breathing, and articulation.

Lower gestures help the singer breath lower therefore the sound is more supported by a good breath and has lots of potential options for phrasing.

The rigidness or flow of the conductor’s hands will affect the tone and volume of the choir. Also, the placement of the hands will alter sound because when conductors conduct lower (waist), singers tend to take deeper breaths.

Gesture is how a conductor conveys emotion to his/her choir. They cannot say what they want during a performance, they can only express it. As a choir we must learn to react to whatever they are trying to communicate. As choir members, we look for that.

If the conductor is like a robot, I find it hard to convey emotion in a piece. I’m not sure where they want me to go with the song.

The gestures command the singer with how loud or soft to sing certain phrases.

Very much so – the smoothness of the gestures informed a more legato feel for the verses, while

91 the sharper motions informed a more accented intonation for the final three notes.

Yes, but only since going to college did I learn any of the intention behind the gestures, from observation and critiques given in Choral Union/Summer Chorale at FSU.

The conductor’s gesture in the video I noticed affect by tone quality and intonation. Also large gesture invites singers to sing louder, and vice versa.

Helps express emotions an intent

His lower gestures reminded me to breath lower, while his higher gestures made me think to breath higher. Also, his speeding up and slowing of tempo.

APPENDIX G

PARTICIPANT VERBATIM COMMENTS (SPECIFIC)

Open-Ended Response Area Following the Question: ‘Did you feel that your singing changed in response to the conductor on the video? If yes, please describe in detail.’

Probably marginally. If the gesture had been more expansive and breaths lower, I probably would have sung more freely. I did crack a little the second time. 

When the conductor raised his gestures, my tone became lighter, and I got louder when he raised his open hands.

I was more tentative on the highest note on a couple different instances because the conductor used a smaller, more precise gesture. I felt myself singing softer and with a breathier tone.

The second two times, the conductor’s hands were higher and his shoulders were more tense. I leaned toward a brighter, more forward tone just naturally, but also tended to forget my breath.

When his gesture was placed higher in his body, I felt the urge to sing softer, take a more shallow breath, etc. Also, his tempo and left hand movement affected my performance.

While watching lower conducting, I felt more apt to sing technically correct, however, higher conducting seemed to influence more lifted pitch to my ear.

The higher gesture energized my singing an brought my eyes to the conductor. The lower gesture helped me take lower breaths. By the fourth time, I knew what was going to be in the gesture, so it wasn’t very stimulating to watch.

Wen the gesture was higher, I felt tension in my larynx. I could feel my tongue raise. When the gesture was lower, I felt that my breath could be freer to move, and I was overall more relaxed.

I was trying to follow his conducting as closely as possible, and it affected my swells, decrescendos, entrances, and cutoffs.

In the last two videos, I feel like I was alert and clear with my singing as opposed to when he was conducting lower.

Once I realized that the conductor’s hands were higher, I sang more legato and paid more attention to the conductor’s phrasing.

The higher he held his hand and larger the gestures opened up the singers making them sing louder, and the opposite is true when he held his hands lower and gestured smaller.

His lower gesture affected my soft palette position – it made me want to sing with a lower breath, soft pallet, and sometimes posture. The opposite is true for his higher gesture.

I feel that my singing changes in general in respect to a conductor. I feel that the higher gesture, however, made me further [unreadable word] t o think about higher pitch.

It was harder to follow the gesture because depending on the video, it was too low or too high. I think the higher gesture helped me a little with intonation.

It seemed that when the conductor was conducting at a higher level (chest, as opposed to waist) it was easier to stay in tune.

When his hands were up closer to his face I was able to get his mood and feel for the piece way better than when they were lower. I had to really try for tempo more when his hands were lower, it was harder to keep in time. When it was up by his face it was a lot clearer for many things like where he wanted us to breath and marking like ritards and word stress/phrasing. Plus his face helped me understand what he wanted overall for the piece.

When his hand were lower I ended up singing softer.

Gesture does affect singing. Your voice reflects the appearance of the conductor.

Just a little conscious change when to breath and ends of phrases

When he was conducting with a lowered ictus I noticed me and the other choristers were having trouble staying in tune.

In life you are always watch the conductor. With that being said that affects the tone and everything that involves singing.

Sometimes the conductor slowed down or he was short and rhythmic so I sang short. Sometimes he showed easy moving gesture so I sang with legato. Yes, the higher the gesture, the more “pingy” it felt. The lower, the move in the throat it felt. Bother were at such extremes, which led to tension.

Maybe I didn’t breath as deeply since I couldn’t see his face when he breathed on the breaks?

It changed because for the first set I tried singing louder than the 2nd set because his gesture was much lower and smaller the 2nd time. He also indicated when to breath and how much to grow/crescendo on the longer notes.

The change in amount and location of movement changed the amount and quality of the sound.

The way he conducted so low made me look down instead of at eye level. However, this melody sits right on my passaggio, so it got to be taxing after a while. Looking down more didn’t help as I felt less connection to my breath.

When his gesture was lower I had to focus harder to see what his hands were doing so I didn’t get off beat with the rest of the group.

I think I did change slightly from video to video, but not very much. I noticed his hands getting lower, and the retardation at the cadence points.

When the conductor on the screen was conducting lower, I found my color was softer, breathier almost, but less accurate.

When the conductor held is hand out, it caused me to keep producing sound and crescendo with vibrato.

I didn’t respond so much the first time, but then when the 8 of us started breathing together, I think we started moving with his gestures as a whole.

The placement of higher hands and more expression o the conductor’s face in the last 2 videos changed my singing to a more deliberate, pianissimo style.

Slightly. I sang softer when the conductor had a high ictus but without variation, I began to go into “auto-singing” mode.

Yes, sometimes he gestured for a crescendo and other he didn’t.

Because of the difference in ictus heights.

When the conductor moved his hands higher, I think I sang with a higher timbre. That also made it easier to watch his face, which made it easier to breath accurately.

When the conductor’s hands were higher, it was easier for me to follow. Also, when he showed a held out note in his left hand, I knew to sustain the note.

When he held his hands below his waist, I naturally wanted to sing quieter. When he kept his hands up and gave a forward gesture, I would crescendo with that gesture. Its harder to sing more musically when his hands were low. I want to see both his hands and his expression.

The first two exercises gave emphasis and I responded well and knew what he wanted. But the second two were relaxed and I felt bored and unengaged.

The conductor gave clear signs of where to break, his eyebrows gave emphasis on specific notes in the phrase, and his hands had differing flaws of beat.

First 2 times felt less grounded in breath than last 2 times. Also, vowels felt darker on the last two times.

At first I was singing through the lines, then I started breaking when he lifted his hands. I noticed when his arms moved to the lower part of his body, but I don’t think I changed my singing.

I get a deeper breath with a lower gesture - less shallow.

I dropped my jaw when the conductor’s hands lowered for his pattern. My sound was more rounded.

Yes. At first I did not pay much attention to his expression other than his happy-looking face, which read “lightly, kindly”, but I certainly noticed when he was conducting at a higher area, using smaller gestures. I then paid attention to his hand (left) and the volume he wanted.

The first conducting was very formal and I sang as I saw, cold and flat. The second set had me

96 growing on long notes and trying to really sing the piece.

I have learned what specific conducting gestures mean so I am able to cooperate with what sound the conductor wants.

When the conductor placed his hand higher in the latter two examples, I felt more alert. This might’ve been because my posture slightly shifted to adjust to looking in a new direction.

When the conductor’s arms were raised I tended to brighten my tone. Left hand gestures created phrases with musicality.

I felt like the higher placement of the ictus didn’t help my singing, although I know we were off tempo with a lower placement. I know it was changing in response, but I can’t accurately describe all of the changes.

I was trying to follow him continuously.

As he extended his hand I tried to make the sound go forward, and if he had withdrawn his hand I would have sang softer gradually.

REFERENCES

Bartholomew, W. (1934). A physical definition of “good voice quality” in the male voice. Journal of the Acoustic Society of America, 6, 25-33.

Bavelas, J. B., Black, A., Lemery, C. R., & Mullett, J. (1986). “I show how you feel”: Motor mimicry as a communicative act. Journal of Personality and Social Psychology, 50, 322-329.

Berlyne, D. E. (1970). Novelty, complexity and hedonic value. Perception and Psychophysics, 8, 279-286.

Bernieri, F. J. (1988). Coordinated movement and rapport in teacher-student interactions. Journal of Nonverbal Behavior, 12, 120-138.

Boersma, P., & Weenink, D. (2012). Praat: Doing phonetics by computer (Version 5.2.35) [Computer program]. Retrieved August 7, 2012, from http://www.praat.org/

Borch, D. Z., Sundberg, J. (2002). Spectral distribution of solo voice and accompaniment in pop music. Logoped Phoniatr Vocol, 27, 37-41.

Bowles, C.L. (1998). Music Activity preferences of elementary students. Journal of Research in Music Education, 46, 193 – 207.

Bretos, J., & Sundberg, J. (2003). Measurements of vibrato parameters in long sustained crescendo notes as sung by ten sopranos. Journal of Voice, 17(3), 343-352.

Brown, S. E. (1991). Determination of location of pitch within a musical vibrato. Bulletin for the Council of Research in Music Education, 108, 15-30.

Brunkan, M. C. (2013). The effects of watching three types of conductor gestures and performing varied gestures along with the conductor on measures of singers’ intonation and tone quality: A pilot study. International Journal of Research in Choral Singing 4(2), 37-51.

Byo, J. L. (1990). Recognition of intensity contrasts in the gestures of beginning conductors. Journal of Research in Music Education, 38, 157-163.

Caclin, A., McAdams, S., Smith, B. K, & Winsberg, S. (2005). Acoustic correlates of timbre space dimensions: A confirmatory study using synthetic tones. Journal of the Acoustical Society of America, 118, 471-482. DOI: http://dx.doi.org/10.1121/1.1929229

Camurri, A., Mazzarino, B., Ricchetti, M., Timmers, R., & Volpe, G. (2004). Multimodal analysis of expressive gesture in music and dance performances. In A. Camurri & G. Volpe (Eds.), Lecture notes in artificial intelligence: Vol. 2915. Gesture- based communication in human-computer interaction (pp. 20-39). Berlin: Springer Verlag.

Carter, B. B. (2007). An acoustic comparison of voice use in solo and choral singing in undergraduate and graduate student singers. Retrieved from ProQuest Digital Dissertations. (AAT 3290821)

Carter, N., Hopkins, J.A., & Dromey, C. (2010). Volitional control of vibrato in trained singers. Journal of Singing, 67(1), 9-17.

Chagnon, R. D. (2001). A comparison of five choral directors’ use of movement to facilitate learning in rehearsals. Retrieved from ProQuest Digital Dissertations. (AAT No. 3002819)

Cofer, R. S. (1994). An investigation into the effects of conducting instruction with seventh-grade band students. Southeastern Journal of Music Education, 6, 107-118.

Cofer, R. S. (1998). Effects of conducting-gesture instruction on seventh-grade band students’ performance response to conducting emblems. Journal of Research in Music Education, 46, 360 – 373.

Coleman, R. F. (1994). Dynamic intensity variations of individual choral singers. Journal of Voice, 8, 196-201.

Conductor. (2014). In Merriam-Webster Online. Retrieved from http://www.merriam-webster.com/dictionary.

Conductor. (2014). The Oxford Companion to Music. Retrieved from http://www.oxfordmusiconline.com

Cook-Koenig, C. (1995). Vocal fatigue in choral singing: Causes and suggestions for prevention voiced by prominent choral directors. Retrieved from ProQuest Digital Dissertations. (AAT 9526487)

Darrow, A. A. & Johnson, C. M. (2009). Preservice music teachers’ and therapists’ nonverbal behaviors and their relationship to perceived rapport. International Journal of Music Education, 27 (3), 269-280.

Daugherty, J. (2001). On the voice: Rethinking how voices work in a choral ensemble. Choral Journal 42(5), 69-75.

Daugherty, J. (2013). Choir spacing and choral sound: Physical, pedagogical and philosophical dimensions. The Phenomenon Of Singing, 2, 77-88. Retrieved from http://journals .library.mun.ca/ojs/index.php/singing/article/view/661

Daugherty, J. F. & Brunkan, M. C. (2013). Monkey see, monkey do? The effect of nonverbal conductor lip rounding on visual and acoustic measures of singers’ lip postures. Journal of Research in Music Education, 60, 345- 362.

Davids, J. & LaTour, S. (2012) Vocal Technique: A Guide for Conductors, Teachers, and Singers. Longgrove, IL. Waveland Press, Inc.

Davidson, J. W. & Correia, J. S. (2002). Body movement. In R. Parncutt & G. E. McPherson (Eds.), The science and psychology of music performance. Creative strategies for teaching and learning (pp. 237-250). New York, New York: Oxford University Press.

Dease, J. P. (2007) The effect of differential feedback on closed- versus open- handed conducting position. Retrieved from ProQuest Digital Dissertations. (AAT 3301541)

Decker, H. & Kirk, C. ( 1995) Choral Conducting: focus on communication. Prospect Heights, Illinois: Waveland Press, Inc.

Demaree, R. W., & Moses, D. V. (1995). The Complete Conductor. Englewood Cliffs, NJ: Prentice Hall.

Detwiler, G. C. (2009). Solo singing technique & choral singing technique in undergraduate vocal performance majors: A pedagogical discussion. Retrieved from ProQuest Digital Dissertations. (AAT 3351947)

Devaney, J. (2011). An empirical study of the influence of musical context on intonation practices in solo singers and SATB ensembles. Retrieved from ProQuest Digital Dissertations. (AAT NR61988)

Diaz, J. A., & Rothman, H. B. (2003). Acoustical comparison between samples of good and poor vibrato in singers. Journal of Voice, 17(2), 179-184.

Droe, K. L. (2006). Music Preference and Music education: A review of literature. Applications of research in music education, 24(2), 23.

Dromey, C., Carter, N., and Hopkin, A. (2003). Vibrato rate adjustment. Journal of Voice, 17(2), 168-178.

Duke, R. A. (1985). Wind instrumentalists’ intonational performance of selected musical intervals. Journal of Research in Music Education, 33, 101- 111.

Ekholm, E. (2000). The effect of singing mode and seating arrangement on choral blend and overall choral sound. Journal of Research in Music Education, 48, 123-135.

Ekholm, E., Papagiannis, G. C., Chagnon, F. P. (1998). Relating objective measurements to expert evaluation of voice quality in Western classical singing: Critical perceptual parameters. Journal of Voice, 12, 182-196.

Elliott, M. (2007). Singing in Style: A guide to vocal performance practices. New Haven, CT: Yale University Press

Fagnan, L. (2005). The acoustical effects of the core principles of the bel canto method on choral singing. Retrieved from ProQuest Digital Dissertations. (AAT NR08601)

100

Farnsworth, P.R. (1950) Musical taste: Its measurement and cultural nature. Stanford: Stanford University Press.

Ferrell M. (2010). Perspectives on Choral and Solo Singing: Enhancing communication between choral conductors and voice teachers. Retrieved from ProQuest Digital Dissertations. (AAT 3434000)

Ford, J. K. (2003). Preferences for strong or weak singer's formant resonance in choral tone quality. International Journal of Research in Choral Singing, 1(1), 29-47.

Ford, J. K. 2001). Implications for non-verbal communication and conducting gesture. The Choral Journal, 42(1), 17-23.

Fuelberth, R. J. V. (2003a). The effect of conducting gesture on singers' perceptions of inappropriate vocal tension: A pilot study. International Journal of Research in Choral Singing, 1(1), 13-21.

Fuelberth, R. J. V. (2003b). The effect of left hand conducting gesture on inappropriate vocal tension in individual singers. Bulletin of the Council for Research in Music Education, 157, 62-70.

Fuelberth, R. J. V. (2004). The effect of various left hand conducting gestures on perceptions of anticipated vocal tension in singers. International Journal of Research in Choral Singing, 2, 27-38.

Fredrickson, W. E., Johnson, C. M., & Robinson, C. R. (1998). The effect of pre- conducting and conducting behaviours on the evaluation of conductor competence. Journal of Band Research, 33(2), 1-13.

Gallops, R. W. (2005) The effect of conducting gesture on expressive- interpretive performance of college music majors. Retrieved from ProQuest Digital Dissertations. (AAT 3188406)

Gambetta, C. (2003). Conducting outside the box: Creating a fresh approach to conducting gesture through the principles of Laban movement analysis. Retrieved from ProQuest Digital Dissertations. (AAT 3201144)

Garnett, L. (2009). Choral conducting and the construction of meaning. Surrey, England: Ashgate.

Garretson, R. L. (1981). Conducting Choral Music. (7th ed.). Boston, MA: Allyn and Bacon, Inc.

Gehrkens, K. (1918). Essentials in conducting. New York: Oliver Ditson Company.

Geringer, J. M. & McManus. (1979). A Survey of Musical Taste in Relationship to Age and Musical Training. College Music Symposium, 19(2), 69-76.

101

Geringer, J. M., MacLeod, R. B., & Allen, M. (2010). Perceived pitch of violin and cello vibrato tones among music majors. Journal of Research in Music Education, 57, 351-363.

Geringer, J. M., MacLeod, R. B., Madsen, C. K., & Napoles, J. (in press). Perception of melodic intonation in performances with and without vibrato. Psychology of Music.

Geringer, J. M., MacLeod, R. B., & Sasanfar, J. K. (2013, March). In tune or out of tune: Are different instruments and voice heard differently? Paper presented at the International Symposium for Research in Music Behavior, Seattle, WA.

Giardiniere, D. C. (1991). Voice matching: A perceptual study of vocal matches, their effect on choral sound, and procedures of inquiry conducted by Weston Noble. Unpublished Doctoral dissertation, New York University.

Ginsborg, J., & King, E. (2009). Gestures and glances: the effects of familiarity and expertise on singers’ and pianists’ bodily movements in ensemble rehearsals. Proceedings of the 7th Triennial Conference of European Society for the Cognitive Sciences of Music Jyväskylä, Finland.

Goodwin, A. W. (1980). An acoustical study of individual voices in choral blend. Journal of Research in Music Education, 28, 119-128.

Goolsby, T. W. (1996). Time use in instrumental rehearsals: A comparison of experienced, novice, and student teachers. Journal of Research in Music Education, 44, 286-303.

Goolsby, T. W. (1999). A comparison of expert and novice music teachers’ preparing identical band compositiongs: An operational replication. Journal of Research in Music Education, 47, 147-187.

Gordon, L. (1989). Choral Director’s Rehearsal and Performance Guide. West Nyack, NY: Parker Publishing Company.

Grady, M.L. (2014, in press). Effects of three conducting gestures (traditional pattern, lateral- only, vertical-only) on acoustic and perceptual measures of choir sound: An exploratory study. International Journal of Research in Choral Singing.

Green, E. A. H. (1987). The Modern Conductor. (4th ed. ). Englewood Cliffs, NJ: Prentice Hall Inc.

Grey, J. M. & Gordon, J. W. (1978). Perceptual effects of spectral modifications on music timbres. Journal of the Acoustical Society of America, 63, 1493-1500. DOI: http://dx.doi.org/10.1121/1.381843

Gumm, A. J. (2012). Six functions of conducting. Music Educators Journal, 99(2), 43

102

Hansen, S., Henderson, A., McCoy, S., Simonson, D., & Smith, B. (2011). Choral directors are from mars and voice teachers are from venus: The top ten complains from both sides of the aisle (or “The farmer and the cowman should be friends”). The Choral Journal, 52(9), 51-58.

Hibbard, T. T. (1994). The use of movement as an instructional technique in choral rehearsals. (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database. (AAT 9418994)

Hicks, V. (1975). Would you enjoy one of your rehearsals? Knowing the four functions of rehearsal language will lead to better communication and a better performance. Music Educators Journal, 62(4), 49-52.

Howard, D. M. (2007). Equal or non-equal temperament in a capella SATB singing. Logopedics Phoniatrics Vocology, 32, 87-94.

Howard, D. M. (2007) Intonation drift in a capella soprano, alto, tenor, bass quartet singing with key modulation. Journal of Voice, 21(3), 300-315.

Howard, D.M., Daffern, H., & Brereton, J. (2011). Quantitative voice quality analyses of a soprano singing early music in three different performance styles. Biomedical Signal Processing and Control, 7, 58-64.

Hung, W.K. & Chen, L. L. (2012). Effects of Novelty and Its Dimensions on Aesthetic Preference in Product Design. International Journal of Design, 6(2).

Hutchins, S. Roquet, C., & Peretz, I. (2012). The vocal generosity effect: How bad can your singing be? Music Perception, 30, 147-159.

Iwarsson, J. (2001). Effects of inhalatory abdominal wall movement on vertical laryngeal position during phonation. Journal of Voice, 15, 384-394.

Jers, H. & Ternstrom, S. (2005). Intonation analysis of a multi-channel choir recording. TMH-QPSR, 47(1), 1-6.

Jordan, J. (1996). A conductor’s gesture: A practical application of Rudolf von Laban’s movement language. GIA.

Jordan, J. & Mehaffey, M. (2001). Choral Ensemble Intonation: Method, procedures & exercises. Chicago, IL: GIA Publications.

Juchniewicz, J. (2008). The influence of physical movement on the perception of musical performance. Psychology of Music, 36(4), 417-427.

Julian, F. (1989). Nonverbal communication: Its application to conducting. The Journal of Band Research, 24(2), 49

Kaplan, A. (1985). Choral Conducting. New York, NY: W. W. Norton & Company, Inc.

103

Kelly, S. N. (1997). Effects of conducting instruction on the musical performance of beginning band students. Journal of Research in Music Education, 45, 295-305.

Killian, J.N. (1985). Operant Preferences for Vocal Balance in Four-Voice Chorales. Journal of Research in Music Education, 32(1), 55-67.

Kirsh, E. R., van Leer, E., Phero, H. J., Xie, C., & Khosia, S. (2013). Factors associated with singers’ perceptions of choral singing well-being. Journal of Voice, 27, 786.e25-786.e32.

Kitch, J. A., Oates, J., & Greenwood, K. (1996). Performance effects on the voices of 10 choral tenors: Acoustic and perceptual findings. Journal of Voice, 10, 217-227.

Knapp, M. L., Hall, J. A., & Horgan, T. G. (2014). Nonverbal Communication in Human Interaction (8th ed). Boston, MA: Wadsworth.

Kotsonis, A. (2014) Are they expressive? The perception of body movement in choral ensembles. Poster presented at the NAfME National Music Research and Teacher Education Conference, St. Louis, MO.

Krause, D. W. (1983). The positive use of conducting stance and motion to affect vocal production and to assist musicality in the training of children’s choirs. Retrieved from ProQuest Digital Dissertations. (AAT 0552760)

Krimphoff, J., McAdams, S., & Winsberg, S. (1994). Characterizing the timbre of complex sounds. II: Acoustic analyses and psychophysical quantification. Journal de Physique, 4, 625-628.

Krone, M. (1949). Expressive conducting. Park Ridge, IL: Neil A. Kjos.

Krudop, D. W. (2003). An examination of the use of kinesics in eliciting expressive ensemble response and its application in an undergraduate choral conducting curriculum. Retrieved from ProQuest Digital Dissertations. (AAT 3137774)

Livingstone, S. T., Thompson, W. F., & Russo, F. A. (2009). Facial expressions and emotional singing: A study of perception and production with motion capture and electromyography. Music Perception: An Interdisciplinary Journal, 26(5), 275-488.

Luck, G., Toiviainen, P., & Thompson, M. R., (2010). Perception of Expression in Conductors’ Gestures: A Continuous Response Study. Music Perception: An Interdisciplinary Journal, 28(1), 47-57.

Madsen, C.K. & Coggiola, J.C. (2001). The effect of manipulating a CRDI dial on the focus of attention of musicians/nonmusicians and perceived aesthetic response. Bulletin of the Council for Research in Music Education, 149, 13-22.

Madsen, C.K. & Geringer, J. M. (1976). Preferences for Trumpet Tone Quality versus Intonation. Bulletin for the Council of Research in Music Education, 46, 13 -22.

104

Madsen, C.K. & Geringer, J. M. (1981). Discrimination between tone quality and intonation in unaccompanied flute/oboe duets. Journal of Research in Music Education, 29(4), 305- 313.

Madsen, C. K. & Geringer, J. M. (2000). A Focus of Attention Model for Meaningful Listening. Bulletin of the Council for Research in Music Education, 147, 103-108.

Madsen, M. K. (1991). The effect of specific conducting gestures on vocal sound (master’s thesis). Florida State University, Tallahassee.

Madsen, M. K. (2009). Effect of aural and visual presentation modes on Argentine and US musicians’ evaluations of conducting and choral performance. International Journal of Music Education, 27, 48 – 58.

Mann, L. M. (In press). The effect of singing mode on vibrato rate, extent and duration of female undergraduate music majors. International Journal of Research in Choral Singing.

Mann, L. M. (2014). The effect of conducting gestures on singer intonation and vibrato. Paper presented at the 2014 NAfME Music Research and Teacher Education National Conference, St. Louis, MO.

Manternach, J. N. (2011). The effect of varied conductor preparatory gestures on singer upper body movement. Journal of Music Teacher Education, 22, 20-34.

Manternach, J. N. (2012). The effect of nonverbal conductor lip rounding and eyebrow lifting on singers’ lip and eyebrow postures: A motion capture study. International Journal of Research in Choral Singing, 4(1), 36-46.

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

Mayne, R. G. (1992). An investigation of the use of facial expression in conjunction with musical conducting gestures and their interpretation by instrumental performers. Dissertation Abstracts International: Section A. Humanities and Social Sciences, 53, 2729.

McClung, A. C. (1996). The relationship between nonverbal communication and conducting: An interview with Rodney Eichenberger. Choral Journal, 36(10), 17-24.

McCoy, S. (2011). The choir issue. The Journal of Singing, 67(3), 297-301.

Miller, R. (1996). The structure of singing: System and art in vocal technique. Belmont, CA: G. Schirmer.

Miller, R. (1997). National schools of singing: English, French, German and Italian techniques of singing revisited. Lanham, MD: Scarecrow Press.

Mitchell, H. F., & Kenny, D. T. (2004). The impact of ‘open throat’ technique on vibrato rate, extent, and onset in classical singing. Logoped Phoniatr Vocol, 29,171-182.

105

Mitchell, H. F., & Kenny, D. T. (2010). Change in vibrato rate and extent during tertiary training in classical singing students. Journal of Voice, 24(4), 427-434.

Morris, R. J., Mustafa, A. J., McCrea, C. R., Fowler, L. P., & Aspaas, C. (2007). Acoustic analysis of the interaction of choral arrangements, musical selection, and microphone location. Journal of Voice, 21, 568-575. Doi:10.1016/j.jvoice.2006.04.006

Morrison, S. J., Price, H. E., Geiger, C. G., & Cornacchio, R. A. (2009). The effect of conductor expressivity on Ensemble Performance Evaluation. Journal of Research in Music Education, 57, 37-49.

Murbe, D., Friedemann, P., Hofmann, G., & Sundberg, J. (2002). Significance of Auditory and Kinesthetic Feedback to Singers’ Pitch Control. Journal of Voice, 16(1), 44-51.

Nair, G. (1999). Voice – Tradition and Technology: A State of the Art Studio. Singular.

Napoles, J. (2007). The effect of duration of teacher talk on the attitude, attentiveness, and performance achievement of high school choral students. Retrieved from ProQuest Digital Dissertations. (AAT 3236752)

Napoles, J. (in press). Verbal instructions and conducting gestures: Examining two modes of communication. Journal of Music Teacher Education.

Napoles, J. & Vazquez-Ramos, A. M. (2013). Perceptions of time spent in teacher talk: A comparison among self-estimates, peer estimates, and actual time. The Journal of Research in Music Education, 60, 452-461.

Nickerson, J. F. (1949). Intonation of solo and ensemble performance of the same melody. The Journal of the Acoustical Society of America, 21(6), 593-595.

Noble, W. (2005). Achieving Choral Blend Through Standing Position. United States, GIA Publications.

Nordmark, J. & Ternstrom, S. (1996). Intonation preferences for major thirds with non-beating ensemble sounds. TMH-QPSR, 37(1), 57-62.

Novak, A., & Vokral, J. (1995). Acoustic parameters for the evaluation of voice for future voice professionals. Folia Phoniatr Logoped, 47, 279-285.

Nowak, J., & Nowak, H. (2002). Conducting the Music, Not the Musicians. New York, NY: Carl Fisher.

Olson, M. (2008). Vibrato vs. non-vibrato: The solo singer in the collegiate choral ensemble. Journal of Singing, 64, 561-564.

Olson, M. (2010). The Solo Singer in the Choral Setting: A handbook for achieving vocal health. Lanham, MD: The Scarecrow Press. Omori, K., Kacker, A., Caroll, L. M., Riley, W. D., Blaugrund, S. M. (1996). Singing power ratio: Quantitiative evaluation of singing voice quality. Journal of Voice, 10, 228-235.

106

Parr, S. M. (1988). The effects of graduated exercise at the piano on the pianist’s cardiac output, forearm, blood flow, heart rate, and blood pressure. Medical problems of performing artists, 3(3), 100.

Powell, S. (1991). Choral Intonation: More than meets the ear. Music Educators Journal, 77(9), 40-43.

Prame, E. (1994). Measurements of the vibrato rate of ten singers. Journal of the Acoustical Society of America, 96, 1979-1984.

Prame, E. (1997). Vibrato extent and intonation in professional western lyric singing. Journal of the Acoustical Society of America, 102(1), 616-621.

Price, H., Morrison, S., Geiger, C., & Cornacchio, R. (2009). The effect of conductor expressivity on ensemble performance evaluation. Journal of Research in Music Education, 57(1), 37-49.

Price, H., & Mann, A. (2011). The effect of conductors on ensemble evaluations. Bulletin of the Council for Research in Music Education, 189(2011): 57-72.

Price, H. E., & Winter, S. (1991). Effect of strict and expressive conducting on performances and opinions of eighth grade students. Journal of Band Research, 27 (1), 30-43.

Reid, K. L. P., Davis, P., Oates, J., Cabrera, D., Ternström, S., Black, M., & Chapman, J. (2007). The acoustic characteristics of professional opera singers performing in chorus versus solo mode. Journal of Voice, 21(1), 35-45. DOI:10.1016/j.jvoice.2005.08.010

Robinson, R., & Winold, A. (1976). The Choral Experience: Literature, Materials, and Methods. Oyster Bay, New York: Harper & Row.

Roe, P. Choral Music Education. Prospect Heights, IL: Waveland Press.

Rossing, T. D., Sundberg, J., & Ternström, S. (1986). Acoustic comparison of voice use in solo and choir singing. Journal of the Acoustical Society of America, 79, 1975-1981.

Rossing, T. D., Sundberg, J., & Ternstrom, S. (1987). Acoustic comparison of soprano solo and choir singing. Journal of the Acoustical Society of America, 82, 830-836.

Sasanfar, J. (2012). Influence of aural and visual expressivity of the accompanist on audience perception of expressivity in collaborative performances of a soloist and pianist. Doctoral Dissertation, Florida State University.

Schubert, E., Wolfe, J., & Tarnopolsky, A. (2004). Spectral centroid and timbre in complex, multiple instrumental timbres. Proceedings of the 8th International Conference on Music Perception & Cognition, (pp. 654-657). Evanston, IL: ICMPC

Schultz, M., & Lipscomb, S. (2007). Hearing gestures, seeing music: Vision influences perceived tone duration. Perception, 36, 888-897.

Seashore, C. E. (1938). Psychology of music. New York: McGraw-Hill.

107

Silvey, B. (2011). The effect of ensemble performance quality on the evaluation of conducting expressivity. Journal of Research in Music Education, 59(2), 162-173.

Silvey, B. (2013). The role of conductor facial expression in students’ evaluation of ensemble expressivity. Journal of Research in Music Education, 60(4), 419-429.

Skadsem, J. (1997). Effect of conductor verbalization, dynamic markings, conductor gesture, and choir dynamic level on singers’ dynamic responses. Journal of Research in Music Education, 45(4), 509-520.

Slusher, H. D. (1991). A comparison of the perspectives of college choir directors, voice teachers, and voice students concerning solo and choral singing. Retrieved from ProQuest Digital Dissertations. (AAT 9130559)

Small, A. R. (1982). The effect of a simultaneous melodic stimulus on harmony intonation of college singers. Psychology of Music, 10, 18-25.

Sousa, G. D. (1988). Musical conducting emblems: An investigation of the use of specific conducting gestures by instrumental conductors and their interpretation by instrumental performers (Doctoral dissertation).

Stauch, T. (1986). An examination of nonverbal communication behaviors of selected collegiate choral directors. Doctoral Dissertation, Arizona State University.

Sundberg, J. (1979). Perception of singing. Speech Transmission Laboratory – Quarterly Progress and Status Report, 20, 1-48.

Sundberg, J. (1982). In tune or not? A study of fundamental frequency in music practice (STL-QPSR 1). Stockholm: Department of Speech Communication, Jngl.

Ternström, S. (1991). Physical and acoustic factors that interact with the singer to produce the choral sound. Journal of Voice, 5, 128-143.

Ternström, S. (1993). Perceptual evaluations of voice scatter in unison choir sounds. Journal of Voice, 7, 129-135.

Ternström, S. (1994). Hearing myself with other: Sound levels in choral performance measured with separation of one’s own voice form the rest of the choir. Journal of Voice, 8, 293- 302. Ternström, S. (1999). Preferred self-to-other ratios in choir singing. Journal of the Acoustical Society of America, 105(6), 3563-3574.

Ternström, S. & Sundberg, J. (1982). Acoustical factors related to pitch precision in choir singing. STL-QPSR, 23(2-3), 76-90.

Titze, I., Bergan, C. C., Hunter, E. J., & Story, B. (2003). Source and filter adjustments affecting the perception of the vocal qualities twang and yawn. Logoped Phoniatr Vocol, 28, 147-155

Tonkinson, S. (1994). The Lombard effect in choral singing. Journal of Voice, 8, 24-29.

108

Vacharkulksemsuk, T. & Frerickson, B. L. (2012). Strangers in sync: Achieving embodied rappyrt through shared movments. Journal of Experimental Social Psychology, 48, 399-402.

Van Besouw, R. M, Brereton, J. S. & Howard, D. M. (2008). Range of Tuning for Tones With and Without Vibrato. Music Perception: An Interdisciplinary Journal, 26(2), 145-155.

Vennard, W. (1967). Singing: The mechanism and the technique. New York, NY: Carl Fischer.

Votaw, L. (1931). Choral Intonation. Music Supervisors’ Journal, 18(1), 50-53.

Wapnick, J. & Ekholm, E. (1997). Expert consensus in solo voice performance evaluation. Journal of Voice, 11, 429-436.

Watkins, R. E. (1986). A descriptive study of high school choral directors’ use of modeling, metaphorical language, and musical/technical language related to student attentiveness. Retrieved from ProQuest Digital Dissertations. (AAT 8618590)

Watts, C., Murphy, J., & Barnes-Burroughs, K. (2003). Pitch matching accuracy of trained singers, untrained subjects with talented singing voices, and untrained subjects with nontalented singing voices in conditions of varying feedback. Journal of Voice, 17(2), 185-194.

Watts, C., Barnes-Burroughs, K., Estis, J., & Blanton, D. (2006). The singing power ratio as an objective measure of singing voice quality in untrained talented and nontalented singers. Journal of Voice, 20, 82-88.

Weber, S. T. (1992). An Investigation of intensity differences between vibrato and straight tone singing. Retrieved from ProQuest Digital Dissertations. (AAT 9223155)

Willard, R. (1986). Speak less, communicate more. The Instrumentalist, 40, 38-41.

Wis, R. M. (1993). Gesture and body movement as physical metaphor to facilitate learning and to enhance musical experience in the choral rehearsal. Retrieved from ProQuest Digital Dissertations. (AAT No. 9327319)

Witt, A. C. (1986). Use of class time and student attentiveness in secondary instrumental rehearsals. Journal of Research in Music Education, 34(1), 34-42.

Yarbrough, C. (1975). Effect of magnitude of conductor behavior on students in selected mixed choruses. Journal of Research in Music Education, 23(2), 134-146.

109

BIOGRAPHICAL SKETCH

Name: Lesley Maxwell Mann Date of Birth: February 1, 1979 Birth Place: Atlantic City, NJ Home Town: Atlanta, GA Higher Education: Florida State University Tallahassee, FL Major: Choral Music Education Degree: B.M.E. (2001)

Florida State University Tallahassee, FL Major: Choral Music Education Degree: M.M.E. (2004)

Florida State University Tallahassee, FL Major: Choral Conducting and Music Education Degree: Ph. D. (2014)

Experience: College of Central Florida Assistant Professor of Vocal Music Ocala, FL (2006 – 2011)

William R. Boone High School Director of Choral Activities Orlando, FL (2001 – 2005)

110