Influencing Oral-Nasal Balance in Speech and Song

INFLUENCING ORAL-NASAL BALANCE IN SPEECH AND SONG

Charlene Holly Santoni

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Faculty of Music University of Toronto

Charlene Holly Santoni Doctor of Philosophy Faculty of Music University of Toronto 2021 Abstract

Oral-nasal balance is dependent on the valving of the velopharyngeal sphincter. This dissertation examined the control and modification of oral-nasal balance in speech and song. Literature was reviewed from the fields of speech science and singing voice pedagogy, which motivated three projects.

The first project (Santoni et al., 2018) explored the role of altered nasal signal level feedback on the regulation of oral-nasal balance in singing in trained and untrained singers. Results indicated that all participants showed lower nasalance scores in response to both increased and decreased nasal signal level feedback. The findings implied that trained singers’ internal models for controlling oral-nasal balance may not be as refined as those that guide other vocal parameters.

The second project (Santoni et al., 2019) explored the influence of voice focus adjustments

(forward and backward vocal tract shape and length modifications) (Boone, 2007; Boone et al.,

2010) on the control of oral-nasal balance in typical speakers in speech and song. Results indicated that forward focus led to higher and backward focus to lower nasalance scores. There was one exception wherein one female participant produced lower nasalance scores in forward focus during the nasal stimulus. The results confirmed that voice focus maneuvers influence oral-nasal balance in typical speakers. Findings from project 2 led to the clinical experiment described in project 3.

The third project (Santoni, Thaut & Bressmann, 2020) explored the influence of voice focus adjustments on the control of oral-nasal balance in speakers with hypernasality in speech and song. Results indicated that for a group of five speakers with hypernasality, forward focus resulted in higher and backward focus resulted in lower nasalance scores. There was one exception wherein one male participant produced lower nasalance scores in forward focus during the nasal stimulus. The results provided preliminary evidence that voice focus maneuvers influence oral-nasal balance in speakers with hypernasality. With further development of the voice focus method, the intervention could become a useful therapeutic approach for speakers with hypernasality.

Keywords: resonance, oral-nasal balance, velopharyngeal sphincter, nasality, hypernasality, cleft palate, nasalance, auditory feedback, compensation, voice focus, chiaroscuro, speech language pathology, singing voice pedagogy, voice therapy.

iii

Preface

This thesis consists of studies that were designed with the support of members of the thesis committee, who contributed their disciplinary, analytical, pedagogical and clinical expertise to assist with the development of questions to which defined each of the three projects that make up this dissertation. Committee members (Dr. Michael Thaut, Dr. Tim Bressmann and Prof. Lorna

MacDonald) reviewed and provided feedback for different parts of this manuscript. The PhD candidate is the primary author and contributor to all three projects, designing objectives and methodology, completing data extraction and analyses, and preparing and revising the manuscripts.

Harmonia: an agreement of sounds.

Acknowledgments

“…the best way to ‘know’ a thing is in the context of another discipline.” (Leonard Bernstein, The Unanswered Question: Six Talks at Harvard, 1973). What an enormous priviledge to have been able to complete an interdisciplinary PhD that brought together the fields of speech science and singing voice pedagogy. I am filled with immense gratitude and indebted to a number of people for making it possible.

First and foremost, to my supervisors, to whom I am eternally grateful. To Dr. Tim Bressmann: I owe so much of my thanks to you. Thank you for believing in the music student with so many enthusiastic ideas about how to bridge singing and speech language pathology. Thank you for the masterful apprenticeship in voice research, for challenging me to step outside of my comfort zone, for our insightful discussions about oral-nasal balance, and for always keeping me focused. Most of all, thank you for sharing what you know with me and for truly caring about my success. To Dr. Michael Thaut. Thank you for your extraordinary wealth of inspirational accomplishment related to translational research in music and rehabilitation. You opened up a world of possibility to me. Thank you for seeing the spark in my eye when I spoke about singing and voice rehabilitation, for nurturing that ambition, for encouraging me to do clinical work, and for always vouching for my abilities. To my PhD committee member and lifelong mentor Prof. Lorna MacDonald. Thank you for revealing the wonderful world of vocal pedagogy and voice science to me; it became a true love of mine. I can still remember poring over that first pedagogy textbook you gave me years ago (the Doscher book) every chance I got. Thank you also for never wavering in your support of my creative and often wildly ambitious ideas, and for always championing my work. To my PhD examiners Dr. Paolo Campisi and Dr. Jamie Perry. Thank you as well for your careful review of my work.

To my additional mentors, all of whom had a hand in my earning this degree. To my beloved voice teacher Dr. Mary Morrison. You are without question, one of the great singing teachers of our time and learning from you has been an honour that I will forever cherish. One of the most profound things you ever said to me was: “I’m proud of you because you took a risk.” Thank you for your wisdom, for your unconditional support and for being the kind of teacher that I will always strive to be. To Dr. Darryl Edwards. Thank you for your invaluable mentorship and for always encouraging me with such heartfelt and unwavering enthusiasm. Thank you most of all vi

for being the person that suggested: “What if you did a PhD?” You changed the course of my life. To Dr. Corene Hurt-Thaut. Thank you for impressing upon me that I had something unique to offer and for giving me my first platform with which to engage in music-inspired voice rehabilitation. On a more personal level, thank you for sharing with me that you too completed a PhD with little children at home and that it was possible! Your support meant so much.

To my labmates at MaHRC and in the Voice and Resonance Lab for their kindred friendship, especially Monique (who patiently taught me how to use all of the lab equipment), Telma, Varsha, Jenni, Catherine, and Nicole. Most importantly, to my very best lab friend Dr. Gillian de Boer. Thank you for always being excited about my work, for sharing your passion for research and statistics with me, for talking through ideas with me, and for reminding me to take care of myself.

Without question, the projects that make up this thesis would not have been possible without all of my study participants, so thank you to each of them as well, especially the children and staff at Five Counties Children’s Centre.

To my family for all of the immeasurable ways in which they have supported and encouraged me throughout my years of study. To my husband Dan (my other half). Your unwavering faith in my abilities is something that I needed in my life and will treasure forever. I can always picture your admiring face and recall your invaluable advice when I need to. Thank you for reminding me that it’s a marathon, not a sprint, and for repeating this phrase to me so often during my years of graduate study: “You can do it, and no matter what, we love you.” I love you too – so very much. To my son Oliver who was such a blessed surprise in the midst of graduate school and who allowed me a peaceful period in my life with which to discern the kind of work I wanted to do. To my son Jonas, our most recent family blessing. Thank you for being such a good napper so I that could finish my revisions, and for being the boost I needed to see this through to the end. To my twin sister Lesley, my partner on this journey. Who knew that during a global pandemic, the twins would both be awarded PhD’s? Lesley, I knew you could do it, and you knew I could do it, and that made all the difference. To my Mom and Dad who were always the first to stand up and yell “Brava” at my concerts and who always had something incredibly motivating to say when I felt discouraged. Thank you for raising a wildly ambitious daughter

vii

with a lot of passion, grit and determination – I share this accomplishment with you. To my grandparents who were always excited to hear about my latest adventures. You would have gotten a real kick out of this. I miss you every day. To the rest of my family and countless friends (too many to name) who checked in over the years to ask how it was going and wished me well. Thank you. Most of all, thank you to God who set out this path for me all along. “I can do all things through Christ who strengthens me.” (Philippians 4:13).

Two closing thoughts: Some years ago, I taught a young girl with an intricate composite of vocal dysfunction to sing. My work with her in fact drove me to this work. Thank you SF, I will forever be indebted to you. Finally, during the course of my graduate study, I had the immense priviledge to meet and converse at a research conference with Dr. Johan Sundberg, the great 21st century singing voice scientist. During our conversation he impressed upon me that in order to advance the field of singing voice science, more quantitative research is needed. I hope that the findings in this dissertation will in some small way add to this important initiative.

viii

Table of Contents

Preface ...... iv Acknowledgments ...... vi List of Tables ...... xii List of Figures ...... xiii List of Appendices ...... xiv Chapter 1 - Introduction ...... 1 1.1 General Introduction ...... 1 1.2 Anatomy ...... 2 1.2.1 Structure ...... 2 1.2.2 Muscular activity ...... 3 1.2.3 Motor innervation ...... 4 1.2.4 Velopharyngeal function ...... 4 1.3 Cleft palate ...... 5 1.4 Acoustic resonance ...... 6 1.4.1 History ...... 6 1.5 Terminology: Resonance and oral-nasal balance ...... 9 1.6 Disordered oral-nasal balance ...... 10 1.6.1 Hypernasality, hyponasality and mixed nasality ...... 10 1.7 Assessment of velopharyngeal structure and function ...... 11 1.7.1 Direct instrumental assessment ...... 11 1.7.2 Indirect instrumental assessment ...... 12 1.7.3 Auditory-perceptual assessment ...... 15 1.8 Surgical and prosthodontic treatment of cleft palate ...... 17 1.8.1 Surgical treatment ...... 17 1.8.2 Prosthodontic treatment ...... 17 1.9 Behavioural treatment of oral-nasal balance disorders ...... 18 1.9.1 The issue of volitional control of the velopharyngeal sphincter ...... 20 Chapter 2 - Control of vocal parameters in speech and song ...... 21 2.1 Feedback and compensation of intensity, frequency and vowel formants ...... 21 2.1.1 Feedback and compensation in speech ...... 21 2.1.2 Feedback and compensation in song ...... 22 2.2 Models of speech motor control ...... 23 2.3 Control of oral-nasal balance in speech ...... 25 Chapter 3 - Control of oral-nasal balance in song ...... 26 3.1 Historical controversy ...... 26 3.1.1 Schools opposed to nasality ...... 26 3.1.2 Schools that emphasized nasality ...... 26 3.2 Misunderstood terminology ...... 27 3.3 Research overview: controlling nasality in singing ...... 31 3.3.1 Velopharyngeal control in singing ...... 31 3.3.2 Acoustic measurement of nasality in singing ...... 35 3.4 Research objective (project 1) ...... 37 Chapter 4 - Influence of altered auditory feedback on oral-nasal balance in song ...... 39 4.1 Abstract ...... 40 4.2 Introduction ...... 41

4.3 Material and methods ...... 45 4.3.1 Participants ...... 45 4.3.2 Stimulus ...... 45 4.3.3 Participant training ...... 46 4.3.4 Recording procedures ...... 46 4.3.5 Data analysis ...... 47 4.4 Results ...... 47 4.5 Discussion ...... 49 4.6 Conclusion ...... 51 4.7 Acknowledgements ...... 51 Chapter 5 - Modification of oral-nasal balance in speech and song ...... 52 5.1 Voice focus in singing: chiaroscuro ...... 52 5.1.1 Chiara: the forward sound ...... 53 5.1.2 Oscuro: the backward sound ...... 54 5.2 Voice focus in speech ...... 56 5.3 Formant tuning ...... 58 5.4 Vocal tract impedance ...... 59 5.5 Research overview ...... 60 5.6 Research objectives (projects 2 & 3) ...... 61 Chapter 6 - Influence of voice focus adjustments on oral-nasal balance in speech and song ...... 63 6.1 Abstract ...... 64 6.2 Introduction ...... 65 6.3 Material and methods ...... 68 6.3.1 Participants ...... 68 6.3.2 Stimuli ...... 68 6.3.3 Participant training ...... 69 6.3.4 Recording procedures ...... 70 6.3.5 Data analyses ...... 70 6.4 Results ...... 71 6.5 Discussion ...... 77 6.6 Conclusion ...... 82 6.7 Acknowledgements ...... 82 Chapter 7 - Immediate effects of voice focus adjustments on hypernasal speakers’ nasalance scores ...... 83 7.1 Abstract ...... 84 7.2 Introduction ...... 85 7.3 Material and methods ...... 89 7.3.1 Participants ...... 89 7.3.2 Stimuli ...... 90 7.3.3 Participant training ...... 91 7.3.4 Recording procedures ...... 92 7.3.5 Data analysis ...... 93 7.4 Results ...... 93 7.5 Discussion ...... 97 7.6 Conclusion ...... 101 7.7 Acknowledgements ...... 101 Chapter 8 - Conclusion ...... 102

8.1 Oral-nasal balance control ...... 102 8.2 Project 1: Influence of altered auditory feedback on oral-nasal balance in song ...... 103 8.2.1 Summary ...... 103 8.2.2 Significance ...... 103 8.2.3 Limitations ...... 104 8.2.4 Future research perspectives ...... 104 8.3 Oral-nasal balance modification ...... 105 8.4 Project 2: Influence of voice focus adjustments on oral-nasal balance in speech and song ...... 106 8.4.1 Summary ...... 106 8.4.2 Significance ...... 107 8.4.3 Limitations ...... 108 8.4.4 Future research perspectives ...... 108 8.5 Project 3: Immediate effects of voice focus adjustments on hypernasal speakers’ nasalance scores ...... 108 8.5.1 Summary ...... 108 8.5.2 Significance ...... 109 8.5.3 Limitations ...... 110 8.5.4 Future research perspectives ...... 111 8.6 Conclusion and future directions ...... 111 References ...... 115 Appendices ...... 149 Copyright Acknowledgements ...... 151

List of Tables

Table 4.1. Mean nasalance scores and standard deviations (SD) for the groups of speakers in the different feedback conditions (N = 30).

Table 6.1. Mean nasalance scores and standard deviations in percent for the stimuli in the different speaking conditions (N = 20).

Table 6.2. Mean frequency measurements and standard deviations in hertz for the speech stimuli in the different speaking conditions (N = 20).

Table 6.3. Mean intensity measurements and standard deviations in uncalibrated decibels for the stimuli in the different speaking conditions. Results are combined for male and female speakers (N = 20).

Table 7.1. Participant information.

Table 7.2. Individual nasalance scores (%) and absolute differences for all of the stimuli in the four voice focus conditions (baseline 1, forward, backward and baseline 2) (N = 7, n = 6).

xii

List of Figures

Figure 3.1. Spectrogram for spoken and sung versions of normal /i/ and nasalized /i/.

Figure 4.1. Singing task.

Figure 4.2. Schematic diagram of equipment for influence of altered auditory feedback on oral- nasal balance in song.

Figure 4.3. Error-bar chart of the average nasalance scores in the different nasal signal level feedback conditions (N = 30).

Figure 5.1. Forward voice focus diagram.

Figure 5.2. Backward voice focus diagram.

Figure 6.1. Singing task (Reproduced from: Santoni, C., de Boer, G., Thaut, M., & Bressmann, T. (2018). Influence of altered auditory feedback on oral-nasal balance in song. Journal of Voice, 34(1), 157.e9-157.e15. doi:10.1016/j.jvoice.2018.06.014).

Figure 6.2. Boxplots of the average nasalance scores for the stimuli in the different speaking conditions (N = 20).

Figure 7.1. Boxplots of the hypernasal group’s mean nasalance scores (%) for the speech stimuli in the four voice focus conditions (baseline 1, forward, backward and baseline 2) (n = 5).

xiii

List of Appendices

Appendix 1.1 & 6.2. Speech stimuli.

Appendix 6.1. Instructional protocol for influence of voice focus adjustments on oral-nasal balance in speech and song.

xiv

Chapter 1 - Introduction 1.1 General Introduction During speaking and singing, the majority of a person’s speech sound comes out of their mouth and some of it comes out of their nose (Kummer, 2014). Balancing oral and nasal sound transmission is regulated by the opening and closing of a muscular valve called the velopharyngeal sphincter (Yanagisawa, Kmucha & Estill, 1990). In order to produce nasal sounds (e.g., /m/, /n/, /ŋ/) the velum is lowered, which opens the velopharyngeal sphincter allowing sound to resonate in the nose. In order to produce oral sounds (e.g., oral consonants and oral vowels), the velum is lifted, which closes the mechanism allowing sound to resonate in the mouth. Overall, a competent velopharyngeal sphincter is essential for the normal execution of both speech and song (Peterson-Falzone et al., 2001)

When the velopharyngeal sphincter is not functioning properly, this results in abnormal sound transmission through the vocal tract filter resulting in an oral-nasal balance disorder. For example, if the velopharyngeal sphincter is unable to close, this results in excessive sound transfer through the nose, causing speech to become hypernasal (Kummer, 2014). Hypernasality is particularly noticeable during the production of oral consonants and oral vowels. Previous studies have emphasized that hypernasality can impact acceptability and intelligibility of speech (Blood & Hyman, 1977; Whitehill, Gotzke & Hodge, 2013). Affected children have been reported as the victims of social ostracism, seen as less likely to get good grades and as having a reduced ability to make friends (Watterson et al., 2013).

Treating hypernasal speech using speech therapy alone is difficult because conscious proprioception in the velopharyngeal sphincter is limited, making the mechanism difficult to control voluntarily (Hixon, Weismer & Hoit, 2008). This leaves speech therapists with limited therapeutic options. Improving understanding of how to influence oral-nasal balance is essential.

Oral-nasal balance is an important concept in singing instruction (Stark, 1999). Examining its regulation in singing could arguably generalize to speech. Respectively, this dissertation describes research that explores the control and modification of oral-nasal balance in speech and singing. The studies presented focus on gaining deeper understanding of the feedback

mechanism that regulates oral-nasal balance, and on investigating the effect of voice focus adjustments on oral-nasal balance.

This introductory chapter describes the anatomy and function of the oral and nasal cavities, and velopharyngeal sphincter. Cleft palate, a craniofacial anomaly causing velopharyngeal insufficiency, is described thereafter. A historical overview of acoustic resonance then follows with particular attention paid to demarcation of the term oral-nasal balance and to its related disorders. Currently available assessment and behavioural treatment options are also reviewed.

1.2 Anatomy 1.2.1 Structure The nasal cavity is responsible for resonating nasal sounds during speech (Hixon, Weismer & Hoit, 2008). The nasal anatomy includes the external nose, the internal nasal cavity and the nasopharynx (Zalzal, O’Brien & Zalzal, 2018). The nasal cavity consists of the nasal septum and the turbinates. The nasal septum divides the nasal cavity into a left and right side. Both sides of the nasal septum contain three shell-shaped turbinates (superior, middle and inferior) that warm, moisten and filter air as it flows through the nose (Dalgorf & Harvey, 2013). The hard palate acts as a partition between the oral and nasal cavities. It forms the bottom of the nasal cavity and the top of the oral cavity (Hopkins, 2016; Perry & Zajac, 2017). The oral cavity is responsible for resonating oral sounds during speech (Hixon, Weismer & Hoit, 2008). The oral anatomy includes the oral cavity and the oropharynx. The oral cavity consists of the lips, alveolar ridges, the tongue, faucial pillars, the hard and soft palate (velum), the jaw and the floor of the mouth. The hard palate comprises the anterior 2/3 of the roof of the mouth and is made up of bone (Shah & Garritano, 2015). It consists of the premaxilla (a small bone that holds the upper front teeth), the paired palatine processes of the maxilla (the bone that forms the upper jaw) and behind it, the paired horizontal plates of the palatine bone. The hard palate is continuous with the velum. The velum comprises the posterior 1/3 of the roof of the mouth, terminating at the uvula (Perry & Zajac, 2017). The velum is made up of muscle that moves up and back to touch the back and sides of the pharyngeal walls (Moore & Persaud, 1998). The pharynx is the passageway for air and sound. It consists of a musculo-membranous tube that sits behind the nasal and oral cavities (i.e., the naso- and oro-pharynx), and above the larynx (i.e., the hypo-pharynx) (Stemple, Glaze & Klaben, 2010).

The velopharyngeal sphincter is located in between the oral and nasal cavities and includes the velum and the sides and back of the pharyngeal walls (Perry, 2011). The mechanism is made up of a group of muscles that work together in a synchronized valve-like fashion (Moon & Kuehn, 2004). The function of the velopharyngeal sphincter is to create a firm seal between the velum and the pharyngeal walls to separate the oral and nasal cavities (Lubker, 1968; Peterson-Falzone et al., 2001). Normal valving is important for ventilation of the middle ear, the prevention of nasal regurgitation, bolus and liquid propulsion during swallowing, non-speech oral-nasal airflow, and the management of airflow through the mouth and nose required for normal speech (Kuehn & Perry, 2009).

1.2.2 Muscular activity Retraction and elevation of the velum, and constriction of the sides and back of the pharyngeal walls close the velopharyngeal sphincter (Perry, 2011). With the exception of the uvular muscle, all of the velopharyngeal muscles are paired and exist on either side of the midline (Moon & Kuehn, 1996). The muscles do not work alone as closure of the mechanism is dependent on their coordinated and synergistic activity (Kuehn & Moller, 2000). The velar muscles associated with speech production include the levator veli palatini, the musculus uvulae, and the palatoglossus. The levator veli palatini is the main velopharyngeal muscle. Its primary function is to elevate the velum by pulling it up and back toward the posterior pharyngeal wall (Kummer, 2014). The musculus uvulae is also a velar elevator. It shortens the velum by bunching it back in position forming the velar eminence (a convex bulge on the nasal surface of the velum) (Kuehn & Moon, 2005). The palatglossus lowers the velum and elevates the posterior of the tongue. Two additional muscles are associated with the velopharyngeal mechanism. The palatopharyngeus is associated with the sphincter-like action of the velopharyngeal sphincter (Cassell & Elkadi, 1995). It lowers the velum and narrows the pharynx via medial movement of the lateral pharyngeal walls towards the midline to assist with swallowing. While the tensor veli palatini is a velopharyngeal tensor muscle that was historically thought to assist with lateral stretching of the velum, it does not contribute to velar elevation or depression, so its role in velopharyngeal closure has been described as minimal (Kummer, 2014). Its main function is to open the Eustachian tube during swallowing and yawning. In addition to the velar muscles, two pharyngeal muscles also contribute to velopharyngeal closure: the superior pharyngeal

constrictor and the salpingopharyngeus. The superior pharyngeal constrictor narrows the lateral pharyngeal walls towards the midline (Kuehn & Perry, 2009). The salpingopharyngeus also contributes to lateral contraction of the pharynx, albeit very mildly (Kuhen & Perry, 2009; Perry, 2011; Perry & Zajac, 2017). Historically, in the speech-language pathology literature, the salpingopharyngeus has been described as not consistently present in the majority of individuals (Trigos, Ysunza, Vargas & Vazquez, 1988), but anatomists have not reported problems detecting it (Sumida, Yamashita & Kitamura, 2012).

1.2.3 Motor innervation Motor innervation of the velopharyngeal sphincter comes from a network of nerve fibers situated along the back wall of the pharynx called the pharyngeal plexus. It consists of branches from the glossopharyngeal (CN-IX) (posterior tongue/pharynx) and vagus (CN-X) (larynx) nerves (Kuehn & Perry, 2009). Innervation of the palatopharyngeus and levator veli palatini has been attributed to CN-IX and CN-X. Innervation of the musculus uvulae and the salpingopharyngeus has also been attributed to CN-X (Broomhead, 1951; 1957; Domenech-Ratto, 1977; Shimokawa, Yi & Tanaka, 2005). In addition, the lesser palatine nerve, which branches off the maxillary nerve, coming from the trigeminal nerve (CN-V) further innervates the levator veli palatini as well as the palatopharyngeus (Shimokawa, Yi & Tanaka, 2005). The facial nerve (CN-VII) has also been suggested as a contributor to motor innervation of the levator veli palatini (Sedlackova, Lastovka & Sram, 1973; Logjes, Bleys & Breugem, 2016). While not directly involved in velopharyngeal closure, innervation of the tensor veli palatini has been attributed to the mandibular branch of CN-V (Shankland, 2001; Logjes, Bleys & Breugem, 2016).

1.2.4 Velopharyngeal function When the velopharyngeal port is open and at rest, the velum lays against the tongue base and occupies a low region of the pharynx. This allows for the unobstructed movement of air between the lungs and nasal passage during normal breathing (Kummer, 2014). When the velopharyngeal port closes, four standard closure patterns have been documented in the literature: coronal, circular, circular with Passavant’s ridge, and sagittal (Croft, Shprintzen & Rakoff, 1981). These patterns are informed by the orientation of the velopharyngeal muscles, which can be highly individualized (Finkelstein et al., 1993). Coronal is the most common pattern wherein dominant movement comes from the velum, which moves backward to meet the posterior pharyngeal wall.

When the pattern is circular, it involves equal movement of the lateral pharyngeal walls and velum. Circular with Passavant’s ridge (a bulge on the posterior pharyngeal wall) is commonly seen in patients with cleft palate. It is analogous to circular, but with the inclusion of movement from the posterior pharyngeal wall. The last and least observed pattern is sagittal, which involves mainly lateral pharyngeal wall movement (Croft, Shprintzen & Rakoff, 1981; Finkelstein et al., 1995; Hartnick & Boseley, 2010).

Overall, velopharyngeal closure patterns are reported as task-dynamic (McWilliams & Bradley, 1965; Flowers & Morris, 1973; Shprintzen et al., 1975; Matsuya, Yamaoka & Miyasaki, 1979; Goldstein, 2002), falling into two main categories: non-pneumatic and pneumatic. Non- pneumatic activities induce exaggerated, forceful closure patterns and include swallowing, gagging and vomiting. Pneumatic activities include those that utilize air pressure (positive or negative), which require milder closure patterning (Perry, 2011). These activities include blowing, whistling, sucking, kissing, speaking and singing. Gender and age have also been reported as contributing factors to velopharyngeal closure configuration. Males have a larger and wider nasopharynx than females and therefore, more pronounced posterior velar elevation is required of them (Kuehn & Moon, 1998). In children, the nasopharynx is quite small and narrow compared to adults, so less closure effort is required of them (Bzoch, 1997).

When the velopharyngeal sphincter is not functioning properly, this can result in structural velopharyngeal insufficiency (Peterson-Falzone et al., 2001). An example disorder characterized by the symptom of velopharyngeal insufficiency is Cleft palate.

1.3 Cleft palate Cleft palate is a craniofacial anomaly that occurs in utero. Between the 4th and 10th weeks of gestation, an atypical embryological development results in the two sides of the palate not joining together causing an orofacial cleft or opening on the roof of the mouth (Peterson- Falzone, Hardin-Jones & Karnell, 2010). This results in structural velopharyngeal insufficiency, causing hypernasality (Zajac & Vallino, 2017). An overt palatal cleft is visibly open and can be observed upon intraoral inspection. A submucous cleft palate may not be obvious upon intraoral inspection. With a submucous cleft palate, the oral mucosa is intact, but the underlying velar musculature has failed to attach at midline (Gosain et al., 1996). According to the World Health

Organization (2020), one in every 700 live births results in an oral cleft [i.e., cleft lip (CL), cleft lip and palate (CL/P), or isolated cleft palate (CPO)] (Mossey & Modell, 2012). Etiology has been documented as multifactorial with both genetics and prenatal environment (e.g., heavy cigarette smoking, alcohol or drug use) constituting influential risk factors (Dixon et al., 2011; Beaty et al., 2011). Epidemiological research has shown that cleft palate may appear as part of a syndrome (e.g., 22q11 Microdeletion Syndrome, Treacher Collins Syndrome, Stickler Syndrome), however 50-70% of cleft palate cases are reported as non-syndromic (Jones, 1988; Mitchell et al., 2002). Cleft palate is surgically treatable. However, post-surgery, affected children may still require speech intervention to monitor their speech development (e.g., children may make the same maladaptive articulation errors that were developed prior to surgery) (Paal et al., 2005). Speech therapy is recommended for children assessed as having a physically capable velopharyngeal sphincter (Kummer, 2014).

Overall, a properly functioning velopharyngeal mechanism plays a central role in the appropriate management of acoustic resonance - a phenomenon with a dynamic history.

1.4 Acoustic resonance 1.4.1 History In physics, the scientific understanding of resonance dates back to the 6th Century BCE, in particular the Greek philosopher Pythagoras (580-500 BCE). Pythagoras described the relationship between musical intervals and the length of strings (e.g., when plucked, a string half the length of another will play a pitch one octave higher) (Lindsay, 1945). Further contribution to the field came from Italian physicist Galileo Galilei (1564-1642) who observed that pitch was not dependent on the length of a string, but on the frequency of its vibrations (Caleon & Ramanathan, 2008). French polymath Marin Mersenne (1588-1648), often referred to as “The Father of Acoustics”, expanded on Galileo’s work about frequency and pitch. Using a rope, Mersenne estimated the frequency of a standing wave based on the rope’s physical characteristics (i.e., its length, tension and mass) (Mersenne & Chapman, 1957; Rossing, 2007). French physicist Joseph Saveur (1653-1716) studied this relationship further. Saveur calculated the frequencies of organ pipes by counting the beats (i.e., undulations) between two low-pitched pipes playing a semitone interval apart. He treated the count as the difference between the frequencies of the two pipes, which allowed him to calculate their respective individual

frequencies (e.g., (Semmens, 1981). Saveur was the first to apply the term acoustics to the science of sound: “I have come then to the opinion that there is a science superior to music, and I call it acoustics” (Rossing, 2007, pp. 14).

The study of pitch evolved into the observation that sound quality is complex. Swiss mathematician Daniel Bernoulli (1700-1782) furthered the work of Mersenne and Saveur, devising an equation for the vibrating string. Bernoulli stated that the motion of a string is subject to all of its characteristic frequencies (e.g., when two waves are in the same medium [i.e., air], their overall amplitude is the sum of the wave amplitudes at that point) (Rossing, 2007). French mathematician Joseph Fourier (1768-1830) expanded on Bernoulli’s work demonstrating mathematically that it is possible to break down any complex periodic wave (frequency) into simpler waves (harmonics), and that that the sum of the simpler waves is equal to the complex wave (Wood & Bowsher, 1965; Levin, 2009).

The study of the acoustics of sound eventually led to study of the acoustics of speech. In 1779, Russian physicist Christian Gottleib Kratzenstein (1723-1795) designed a set of five acoustic resonators made out of pipe that, when activated by a vibrating reed, sounded like vowels. Similarly, in 1791, Hungarian inventor Wolfgang Von Kemplen (1734-1804) constructed a manually operated speaking machine patterned after the human vocal tract that consisted of an accordion-like bellow (i.e., the lungs) that provided air to sound a reed (i.e., the vocal folds) connected to a hand-held leather resonating tube (i.e., the vocal tract) that could be manipulated to produce whole words and phrases (Schroeder, 1993). While initially dismissed, the machine was later redesigned by Charles Wheatstone (1802-1875) in 1837, garnering better reception (Flanagan, 1972). In 1838, Robert Willis (1800-1875) also used tube-like resonators that emphasized the connection between vowel sounds and vocal tract dimensions. His work demonstrated that vowel quality is dependent on the length of a tube (i.e., the vocal tract), as opposed to its diameter (Schroeder, 1993).

In 1848, Johannes Müller (1801-1858) took the study of simulated speech a step further. Müller analyzed excised larynges, discovering that he could only recreate speech-like sounds when he placed tubes (representing the vocal tract) between the larynx and lips (Müller, 1848). Hermann von Helmholtz (1821-1894) who worked with Müller synthesized this knowledge further. Using

Helmholtz resonators, which were spherical containers with a specific volume and a small opening at one end, Helmholtz established that resonance is determined by the shape of its resonating cavity (i.e., the shape of vocal tract). In Helmholtz’s publication “On the Sensations of Tone” (Helmholtz, 1885), he summarized his findings explaining that while the pitch of a tone is due to its fundamental frequency, the quality of a tone is due to the presence of upper partials or overtones, and that the frequencies of those overtones were multiples of the fundamental (Stark, 1999; Roberts, 2016).

English mathematician Alexander Ellis (1814-1890) helped to move acoustic phonetics towards the study of articulatory phonetics. Ellis proposed that resonance is not only dependent on resonant cavity shaping, but also on the way sounds exits the resonator via the manipulation of the articulators, which allow for the production of consonants and vowels (Ellis, 1877; Stark, 1999). Thereafter, German speech scientist Ludimar Hermann (1848-1914) used an Edison Phonograph to study the spectral properties of consonants and vowels. He observed that vowels peak at resonant frequencies that are characteristic to the vowel, irrespective of pitch, which led him to coin the term “formant” (Weihs et al., 2016).

Swedish speech scientist Gunnar Fant (1919-2009) consolidated the knowledge related to speech acoustics and articulatory phonetics, and developed the source-filter model for speech. The model describes speech as a two-step process wherein the source (i.e., the vibrating vocal folds) produces a harmonic series that is filtered by the resonant properties of the vocal tract (i.e., the pharynx, oral cavity and nasal cavity), which emphasizes certain frequencies and dampens others (Rossing, 2007). Fant’s model is commonly known as the acoustic theory of speech production (Fant, 1960).

Swedish music acoustician Johan Sundberg (b. 1936) built upon Fant’s work. He described the acoustics of the singing voice as determined by the same two factors: the voice source (i.e., the vibrating vocal folds), and the shaping of the vocal tract (i.e., the pharynx, oral and nasal cavities), and refined observations about harmonics and formants (Sundberg, 1998). He described that harmonics are produced by the voice source and change relative to vocal fold shape (i.e., more closure creates stronger and higher harmonics), while formants represent an acoustic phenomenon wherein air inside the vocal tract vibrates at different pitches dependent on

its size and shape (e.g., large areas vibrate at lower pitches and small areas vibrate at higher pitches), selectively emphasizing certain harmonics within a spectral envelope. Sundberg explained that harmonics are quantified as whole number multiples of the fundamental frequency or pitch being sung (e.g., for A3, 1f0 = 220Hz, 2f0 = 440Hz, 3f0 = 660Hz, 4f0 = 880Hz, etc.), whereas formants are represented in terms of frequency spectra. Formant 1 (280-710Hz) and formant 2 (880-2250Hz) are labelled the vowel formants, while a clustering of formants 3-5 make up the so-called singer’s formant (2800-3400Hz), which is a hallmark of unamplified Western classical singing (Sundberg, 1987; Sundberg et al., 1989; Bozeman, 2013).

1.5 Terminology: Resonance and oral-nasal balance Despite its well-established history, acoustic resonance as an attribute of voice is currently not well understood. Current qualification regarding the meaning of the term is somewhat indeterminate between the disciplines of singing voice pedagogy and speech language pathology (de Boer & Bressmann, 2015).

According to, McCoy (2004) resonance as an aspect of singing exists in two forms: sympathetic and conductive. Sympathetic resonance, also termed free or public resonance, requires a resonator. It is dependent on soundwave reverberation within the hollow void of the vocal tract and built upon the relationship of harmonics and formants, as similarly described by Fant (1960) and Sundberg (1998). Conductive resonance, also termed forced or private resonance, does not require a resonator. It refers to soundwaves that are in physical contact with a vibrating body (e.g., bone, cartilage and muscle), often felt across the head and chest. Yet historical pedagogical opinion on how to manage resonance in singing has not been universal (Stark, 1999), largely liable to ambiguous instructional language (e.g., “place the tone forward,” “sing in the mask”, “brighten the sound”, “find ring in the voice”) (Malde, 2009).

According to Peterson-Falzone, Hardin-Jones and Karnell (2001), resonance as an aspect of speech is also not clearly defined. Specifically, speech-language pathologists tend to use the term resonance when referring to nasality and nasality disorders, without clearly distinguishing these from other aspects of speech resonance. De Boer and Bressmann (2015) explained further. Currently, resonance disorders include those related to nasal sound transmission including hypernasality (too much nasal transmission, e.g., velopharyngeal insufficiency), hyponasality

(too little nasal transmission due to a blockage, e.g., deviated septum) and mixed nasality (a mix of hyper- and hypo-nasality). They also include those indicative of sound being trapped in a blind cavity of the vocal tract (nasal, oral or pharyngeal) due to obstruction, typically at the cavity’s exit point (i.e., cul-de-sac resonance). This results in a muffled sound (McWilliams, Morris & Shelton, 1990; Peterson-Falzone, Hardin-Jones & Karnell, 2001; Kummer 2014; Zajac & Vallino, 2017). De Boer and Bressmann (2015) proposed that this can blur assessment because it blends the perceptual impressions of oral-nasal balance with other resonance features. Correctively, they argued that oral and nasal sound transmission be thought of as a two- dimensional space, which can be measured quantitatively, and that all co-occurring auditory- perceptual features of resonance be observed separately as accompanying characteristics. Therefore, the term oral-nasal balance was promoted as a means of clarifying the diagnostic and therapeutic procedures related specifically to oral and nasal sound transmission through the oral and nasal cavities (de Boer & Bressmann, 2015).

1.6 Disordered oral-nasal balance When there is insufficient separation between the oral and nasal cavities, the result is abnormal sound transmission through the vocal tract filter, resulting in an oral-nasal balance disorder (de Boer & Bressmann, 2015; 2016). Oral-nasal balance disorders can stem from velopharyngeal insufficiency (e.g., cleft palate, short velum, enlarged tonsils), velopharyngeal incompetence (e.g., velopharyngeal hypotonia or paralysis, dysarthria, apraxia) or velopharyngeal mislearning (e.g., compensatory misarticulation) (Duffy, 2012; Kummer, 2014; Zajac & Vallino, 2017).

1.6.1 Hypernasality, hyponasality and mixed nasality According to de Boer and Bressmann (2015; 2016), hypernasality, hyponasality and mixed nasality exemplify the three variations of oral-nasal balance disorders. Hypernasality occurs when the velum is unable to elevate due to velopharyngeal dysfunction or oral-nasal fistula (an abnormal opening between the oral and nasal cavity). This makes it impossible for the velopharyngeal sphincter to adduct, which causes too much sound transfer through the nasal cavity, affecting oral speech sounds (vowels and vocalic consonants, i.e., /l/, /m/, /n/ and /r/) (Kummer, 2014). Hypernasality often co-occurs with nasal emission, nasal turbulence, reduced vocal intensity, and facial grimace (Lewis et al., 1993). Cleft palate is commonly characterized by the symptom of hypernasal speech. Hyponasality occurs when there is a posterior obstruction

(e.g., enlarged tonsils) that impedes nasal sound transmission, or when there is restricted space in the nasal cavity or depth in the nasopharynx (e.g., maxillary retrusion) (Kummer, 2014). This results in too little sound energy in the nasal cavity, which affects the production of nasal sounds (e.g., /m/ becomes /b/, /n/ becomes /d/, and /ŋ/ becomes /g/). Mixed nasality constitutes some combination of hyper-and hypo-nasality wherein excessive sound is flowing through the nose due to velopharyngeal dysfunction, but that sound is blocked due to obstruction. Mixed nasality may occur with or without nasal air emission (de Boer & Bressmann, 2015; 2016). Comparatively, hypernasality is the most frequently diagnosed oral-nasal balance disorder in children (Zajac & Vallino, 2017) and has been shown to affect intelligibility and acceptability more significantly than hyponasality (Shprintzen, Lewin & Croft, 1979; Whitehill, Gotzke & Hodge, 2013).

To properly address hypernasality, comprehensive evaluation of an individual’s ability to achieve adequate closure of the velopharyngeal sphincter is a primary first step.

1.7 Assessment of velopharyngeal structure and function Assessment of velopharyngeal structure and function can be done using both direct and indirect methods (Shprintzen, 2013). Direct assessments provide an objective visual evaluation of the velopharyngeal closing mechanism during speech (gap, size and shape) and provide information relative to the articulatory structure of the velopharynx (Lubker & Moll, 1965). Indirect assessments (e.g., aerodynamic, acoustic, auditory-perceptual) are less invasive than direct assessments and provide information that allow clinicians to make inferences regarding velopharyngeal function and help to provide rationales for diagnostic decision-making.

1.7.1 Direct instrumental assessment Direct visual assessments include multi-view videofluoroscopy, nasopharyngoscopy and magnetic resonance imaging (MRI) (Kuehn & Moller, 2000). Computed tomography (CT scan) and lateral radiograph (x-ray) are also commonly used (Bettens, Wuyts & Van Lierde, 2014). With videofluoroscopy, a high-contrast liquid such as a barium sulfate suspension is introduced into the nasal cavity as well as swallowed, coating and improving the contrast of the myomucosal structures in the vocal tract. Thereafter, a real-time x-ray video called fluoroscopy allows for the direct physiologic visualization of all of the components of velopharyngeal

movement (Zajac & Vallino, 2017). Videofluoroscopy is generally the ideal method of assessment for children who are unable to undergo nasopharyngoscopy (Karnell, 2011). With nasopharyngoscopy, a flexible endoscope with a fiberoptic camera and light source is inserted into the nose, through one nostril and into the pharynx. Nasopharyngoscopy provides direct anatomic visualization of the structure and function of the velopharynx during speech (Shprintzen, 2013). While compliance can be problematic for children, since the method is done without radiation and provides the most comprehensive picture of the velopharynx, it is preferred by surgeons (Kummer, 2014). With MRI, magnetic forces and radio waves are used to generate a three-dimensional computer image of all of the velopharyngeal structures (Perry et al., 2014). However, while MRI provides excellent imaging of the velopharyngeal anatomy, it is very costly and therefore not widely used in clinical practice (Perry et al., 2018). CT scan and x-ray have also been used to visualize the velopharyngeal structures, but the images retrieved are static and two-dimensional, which limits the information they provide. Ultrasound has also been used for visualization of the lateral pharyngeal walls, but interpretation of the findings has been considered difficult and subjective (Bettens et al., 2014).

1.7.2 Indirect instrumental assessment Indirect assessments include instrumental aerodynamic and acoustic tools, as well as auditory- perceptual tools. Aerodynamic tests measure nasal air pressure and nasal airflow (Warren, 1967); which is important because fundamentally, the velopharyngeal sphincter functions like an aeromechanical valve that separates the oral and nasal cavities (Zajac & Mayo, 1996; Zajac, 2012). Aerodynamic tests are useful because incomplete closure of the velopharyngeal mechanism can cause symptomatic nasal air emission, which can impair consonant production (Kummer, 2014). Aerodynamic assessment tools include the Czermak nasal mirror, a See-Scape device (ProEd, Austin, Tx), a pneumotachograph (Vitalograph, Lenexa, KS) and an aerophonoscope (Aéro RD, Orqual, France). Specialized versions of the aerophonoscope include the Palatal Efficiency Rating Computed Instantaneously-Speech Aeromechanics Research System (PERCI-SARS, MicroTronics Corp., Chapel Hill, NC), and the Super Nasal Oral Ratiometry System (SNORS, Sharp et al., 1999) (Rose Medical, Canterbury, UK). A Czermak nasal mirror fogging test involves placing a mirror under a patient’s nose to detect nasal air emission. Stethoscopes, listening tubes, air paddles and straws have been used similarly (Lass & Pannbacker, 2015). Errors detected using these methods usually occur on sibilants and affricates

(i.e., turbulent sounds produced when air is forced between the tip of the tongue and the alveolar ridge or palate; e.g., /s/, /z/, /ʃ/, /ʒ/, /tʃ/ and /dʒ/) and plosives such as the lingual-alveolar plosives (i.e., sounds produced with the tip of the tongue at the alveolar ridge; e.g., /t/ and /d/) (Kummer, 2014). A See-Scape device can also detect nasal air emission during speech. The device consists of a rigid plastic tube containing a small styrofoam float that is inserted into a patient’s nose. Nasal air emission causes the float to rise inside of the tube (Zajac & Vallino, 2017). A pneumotachograph measures air flow via a nasal flow tube and air pressure via nasal and oral catheters connected to pressure transducers. These transducers convert the flow and pressure data into electrical signals in order to establish the rate of airflow (Kummer, 2014). An aerophonoscope consists of a handpiece with three airflow sensors placed on the nostrils and mouth that allow for the simultaneous measurement of oral and nasal airflow (Devani, Watts & Markus, 1999). Similarly, the PERCI-SARS also utilizes oral and nasal catheters to measure oral and nasal air pressure and flow. The device’s software interface uses this data to calculate pressure differences between the mouth and nose, and velopharyngeal port size estimates (Sweeney, 2011). Finally, the SNORS consists of a dual chamber mask containing airflow sensors and microphones that extract both aerodynamic and acoustic data. The device registers the amount of nasal airflow as a percentage of total airflow (Bettens, Wuyts & Van Lierde, 2014).

Acoustic spectral analysis is not commonly used to assess oral-nasal balance in clinical settings but has mainly been used in research (Lee, Ciocca & Whitehill, 2004; de Boer & Bressmann, 2016a; Kataoka et al., 1996; 2001). Exceptions include use of the voice low tone to high tone ratio (VLHR) (Lee, Yang & Kuo, 2003) and the multidimensional Nasality Severity Index (Van Lierde et al., 2007; Bettens et al., 2016). Instead, acoustic tests assessing the proportionality of nasal versus oral energy in speech are more the standard. Standard tests include an accelerometer, which measures oral to nasal sound vibration ratios, and devices that measure oral to nasal acoustic energy ratios like the Nasal View (Tiger Electronics), the Oronasal System (Glottal Enterprises), as well as the most popular clinical tool, the Nasometer (KayPentax, Montvale, NJ) (Bettens, Wuyts & Van Lierde, 2014).

1.7.2.1 Nasometry The Nasometer is a diagnostic tool that quantifies nasality during speech by measuring the proportion of oral and nasal acoustic energy (Kummer, 2014). It was first developed in 1970 by Samuel Fletcher, at which time it was called TONAR: The Oral-Nasal Acoustic Ratio (Fletcher & Bishop, 1970). Fletcher subsequently released a TONAR II model in 1976, but its readings were somewhat unreliable. The Nasometer 6200 was introduced by the Kay Elemetrics Corporation in 1986 (Fletcher, 1970; Fletcher, Adams & McCutcheon, 1989) and a second version of the device was released in 2002. The third generation Nasometer II 6450 includes a headset with two directional microphones on either side of a sound separator plate. The headset is secured perpendicular to the face with placement of the plate between a person’s upper lip and nose. The Nasometer software quantifies the relative contribution of the nasal signal to speech extracted every 8 milliseconds, resulting in the calculation of a mean nasalance score for a speech sample. The nasalance score is calculated using the formula: nasalance (%) = nasal/(nasal + oral) x 100 (Fletcher, 1970). Higher nasalance scores indicate more nasality, and lower nasalance scores indicate less nasality (Kummer, 2014).

Standard speech stimuli for nasometric evaluation include: (1) the phonetically balanced Rainbow Passage, used to detect errors during nasal to oral transitions (Fairbanks, 1960); (2) the Zoo Passage; which is loaded with oral phonemes, and used to detect hypernasality; and (3) nasal sentences like “mama made some lemon jam”, used to detect hyponasality (Fletcher, 1976; McWilliams & Philips, 1979) (see Appendix 1.1). Data published in the Nasometer II 6450 manual indicates nasometric normative values for each of these stimuli as follows: (1) Rainbow Passage (31.47%, SD 6.65); (2) Zoo Passage (11.25%, SD 5.63); and nasal sentences (59.55%, SD 7.96) (Kummer, 2005). For the oral stimulus, scoring <20% indicates no hypernasality, scoring between 30-40% indicates mild hypernasality, and scoring >40% indicates substantial hypernasality (Smith & Kuehn, 2007; Kummer, 2014). Dalston, Neiman and Gonzalez-Landa (1993) further proposed an oral passage nasalance cut-off value of ≥28% as a robust indicator of hypernasality, while Smith and Kuehn (2007) proposed that scoring <50% during production of a nasal sentence was a strong indicator of hyponasality.

Specific to children, the Simplified Nasometric Assessment Procedures SNAP Test (MacKay & Kummer, 1994; Kummer, 2005) is a commonly used assessment tool. The test consists of a

series of simple oral, nasal and balanced sentences combined with pictures designed specifically for children that may have trouble with literacy or compliance. For children aged 5-12 years, Fletcher, Adams and McCutcheon (1989) proposed nasometric normative values for standard speech stimuli as follows: (1) Rainbow Passage (35.69%); (2) Zoo Passage (15.53%); and nasal sentences (61.01%). Overall, nasometric assessment of nasality has significant clinical application, and provides a helpful adjunct for the auditory-perceptual analysis of oral-nasal balance (Kummer, 2014).

1.7.3 Auditory-perceptual assessment Indirect auditory-perceptual analysis is one of the most important tools for the speech language pathologist assessing velopharyngeal function (Smith & Kuehn, 2007; Kummer et al., 2012). Standard auditory-perceptual techniques include listening for: (1) oral-nasal imbalances; (2) weak production of oral pressure consonants; (3) audible or turbulent nasal air emission; (4) compensatory misarticulated phonemes (Kummer, 2011); (5) vocal dysfunction symptoms (i.e., dysphonia) (Lewis et al., 1993); as well as (6) poor intelligibility and acceptability (Whitehill, Gotzke & Hodge, 2013).

Several well-established assessment tasks exist. A review of Kummer (2014) and Zajac and Vallino (2017) established this list: (1) sustained phonation (e.g., sustained /s/ to determine utterance length because nasal air emission might reduce a patient’s capacity to sustain phonation; sustained /a/ to detect hypernasality; and sustained /m/ to detect hyponasality); (2) syllable repetition (e.g., /sa sa sa/ to detect nasal air emission; /pa pa pa/ or /ba ba ba/ to detect hypernasality; and /ma ma ma/ or /na na na/ to detect hyponasality or nasal obstruction); (3) counting (e.g., 1-10, 60-70 or 80-90 for hypernasality; and 90-99 for hyponasality); (4) sentence elicitation (e.g., “put the baby in the buggy”, which is loaded with bilabial plosives and aimed at detecting nasal air emission and hypernasality; and “Mama made lemon jam”, which is full of nasals and aimed at detecting hyponasality); (5) the cul-de-sac test, which involves occluding and unoccluding the nostrils to uncover oral-nasal imbalances; and (6) connected speech because it increases the load on the velopharyngeal sphincter, which may contribute to increased hypernasality or nasal air escape (Kummer, 2014; Zajac & Vallino, 2017). Clinicians may also feel the sides of the nose to detect nasal air escape.

Auditory-perceptual assessment is considered the gold standard for a speech language pathologist’s assessment of oral-nasal balance (Kent, 1996; Whitehill & Lee, 2008; Moon, 2009). However, several authors advise that this kind of assessment is problematic. This is due to the fact that auditory-perceptual assessment can be difficult, subjective and has poor intra-rater (1 rater assessing a speech sample more than once) and inter-rater (≥2 raters assessing a speech sample) reliability (Kuehn & Moller, 2000; Whitehill & Lee, 2008; Keuning, Wieneke & Dejonckere, 2004).

Specific to hypernasal speech, distinguishing normal from abnormal speech has been documented as particularly difficult to judge (Watterson et al., 2007). Therefore, equal appearing interval and visual analogue nasality rating scales are commonly used for assessing hypernasality (Yamashita et al., 2018). Scales suggested for clinical practice include: (1) direct magnitude estimation (DME) (Zraick & Liss, 2000; Whitehill, Lee & Chun, 2002; Baylis, Munson & Moller, 2011), (2) the visual analogue scale (VAS) (Svensson, 2000; Baylis, Chapman & Whitehill, 2015), (3) the Visual Sort and Rate (VISOR) method (Granqvist, 2003), and (4) the Borg centiMax scale (Borg, 1982; 2007; Borg & Borg, 2001). More comprehensive standardized clinical outcome tools specific to the cleft population have also been recommended (Bruneel et al., 2020). These tools include: (1) the Great Ormond Street Speech Assessment (GOS.SP.ASS) (Sell, Harding & Grunwell, 1999), (2) the Universal Parameters for Reporting Speech Outcomes in Individuals with Cleft Palate (UPS) (Henningsson et al., 2008), (3) the Cleft Audit Protocol for Speech (CAPS) (Harding, Harland & Razzel, 1997), along with its variations: the Cleft Audit Protocol for Speech-Augmented (CAPS-A) (John et al., 2006, Sell et al., 2009) and the Cleft Audit Protocol for Speech-Augmented Americleft Modification (CAPS-A-AM) (Chapman et al., 2016), and (4) the 2-step protocol designed for the assessment of hypernasality from the Scandcleft project (Lohmander et al., 2009; Lohmander et al., 2017; Willadsen et al., 2017). Each of the tests include single words, sentences and spontaneous speech stimuli that are used for the perceptual judgment of oral-nasal balance, nasal air emission and misarticulation. A standardized set of outcome measures assessing the speech, articulation, and velopharyngeal competence of cleft palate patients was also recently proposed by the International Consortium for Health Outcomes Measurement (ICHOM). Eight main outcomes domains were chosen, which include: (1) eating and drinking; (2) dental and oral health, (3) speech/communication; (4)

otologic health (anatomy and physiology of the ear); (5) breathing; (6) appearance; (7) emotional and psychosocial development; and evaluation of the (8) burden of care (Allori et al., 2017).

Comprehensively, all of the measurement tools listed help clinicians determine appropriate treatment strategies that address velopharyngeal dysfunction, a condition regularly associated with cleft palate.

1.8 Surgical and prosthodontic treatment of cleft palate Cleft palate is commonly treated within comprehensive cleft centers via an interdisciplinary team of professionals combining their expertise in order to maximize treatment success. The team approach encourages treatment planning centered around patient goals to ensure continuity and improved quality of care (Kummer, 2014). These teams usually consist of an oral and plastic surgeon, otolaryngologist, prosthodontist, orthodontist, audiologist, as well as a speech language pathologist (Kuehn & Moller, 2000; Patel et al., 2013). Treatment is also offered via community-based programs outside of the center proper.

1.8.1 Surgical treatment Surgical intervention to repair cleft palate is generally done around 1 year of age (Zajac & Vallino, 2017). However, it has been reported that about 1/3 of patients will continue to exhibit hypernasal speech after surgery (Billmire, 2008) due to insufficient velar lengthening (i.e., muscle tissue deficiency), or faulty insertion of the palatal muscles, specifically the levator veli palatini (Witt & D’Antonio, 1993; Gart & Gossain, 2014). Secondary surgical options mainly include the creation of a pharyngeal flap (i.e., transferring tissue from the back of the pharynx to the palate), which reduces space in the nasopharynx by narrowing the nasal opening, or doing a Furlow Z-plasty of the palatal mucosa, which increases the velar length of an otherwise short palate (Gart & Gosain, 2014; Idle & Monaghan, 2016). Overall, surgical management is of critical importance for providing the structural prerequisites for adequate oral-nasal balance in speech (Zajac & Vallino, 2017).

1.8.2 Prosthodontic treatment Prosthodontic devices may be prescribed when surgery is not possible for a patient, or in cases wherein hypernasality has remained post-surgical intervention (Reisberg, 2000). Standard 17

appliances include a palatal lift prosthesis, which helps to prop up an already adequately long velum, and a speech bulb, which is a palatal prosthesis with an extension that protrudes into the pharynx to help separate the oro- and naso-pharynx. Since these devices may cause discomfort and have been known to stimulate gagging, their effectiveness is highly individualized (Kummer, 2014). A palatal obturator is also commonly recommended (Kuehn & Moller, 2000). The device fills a palatal defect (e.g. fistula) by sitting inside the opening. While not as common, a nasal obturator may also be recommended. The appliance, which has a one-way valve, is placed inside the nose. It reduces exhaled air during speech by redirecting it towards the mouth (Kuehn & Moller, 2000; Leeper, Charles & Sills, 2008). Nasal obturators have been shown to mildly improve intelligibility, but are recommended to be used with caution as they have also been shown to cause hyponasality (Suwaki et al., 2008).

1.9 Behavioural treatment of oral-nasal balance disorders When hypernasality is mild or phoneme-specific (e.g., substituting a nasal for an oral sound), and a patient is physically capable of achieving velopharyngeal closure (McWilliams, Morris & Shelton, 1990), speech therapy is recommended. Speech intervention is typically prescribed post-cleft closure in about 50% of children due to the carryover of pre-surgical compensatory articulation errors (Hardin-Jones & Jones, 2005).

Techniques to reduce hypernasality include perceptual training (Peterson-Falzone et al., 2006), pitch and loudness modifications (Watterson, York & McFarlane, 1994; McHenry, 1997; Boone et al., 2010), altering tongue positioning (Boone et al., 2010), reducing speech rate (Peterson- Falzone, Hardin-Jones & Karnell, 2010), as well as cul-de-sac therapy (i.e., (Morley, 1970; Kummer, 2014) and phonetic articulation therapy (Ruscello, Shuster & Sandwisch, 1991; Golding-Kushner, 1995; Stemple, 2000; Kummer, 2011). Cul-de-sac therapy (i.e., the cul-de-sac technique) involves producing pressure-sensitive sounds with the nose plugged to eliminate nasal emission, and then unplugging the nose in an attempt to produce the sound in the same way (Kummer & Lee, 1996). Phonetic articulation therapy involves establishing new motor speech patterns for vowels and consonants that address active (e.g., developmental substitutions, distortions or omissions) and passive (i.e., structurally-related nasal air emission, nasal turbulence or weak pressure-sensitive consonants) speech sound errors (Nagarajan, Savitha & Subramaniyan, 2009; Kummer, 2014).

Several authors have also mused about oral sound redirection via increased mouth opening (Coston, 1986; McDonald & Koepp-Baker, 1951; Shelton, Hahn & Morris, 1968; Van Demark & Hardin, 1990; Boone & McFarlane, 1994; Peterson-Falzone, Hardin-Jones & Karnell, 2001) and yawning (Kummer, 2014). Non-speech vocal play aimed at detecting a patient’s ability to produce sound with reduced nasality (e.g., singing) (Picheny, Durlach & Braida, 1989; Stemple, Glaze & Klaben, 2010), has also been recommended. As a general rule, using oral-motor exercises (OME’s) (e.g., sucking, blowing, whistling) to strengthen the muscles of the velopharyngeal sphincter has been largely discouraged (Golding-Kushner, 2001; Lof, 2008; Kummer, 2011; McCauley et al., 2009; Ruscello, 2008a, 2008b; Watson & Lof, 2008; Kummer, 2014). This is because research has shown that positive transfer from one behaviour to another requires that the tasks be identical (Watson & Lof, 2008). Since velopharyngeal closure patterns in speech compared to non-speech are different, this would make oral-motor exercises ineffective. This has been documented by several authors (Flowers & Morris, 1973; Shprintzen et al., 1974; Golding-Kushner, 2001; Peterson-Falzone et al., 2006). OME’s are also not recommended to patients with structural abnormalities devoid of a physically capable velopharyngeal closure mechanism (Watson & Lof, 2008; Kummer, 2014).

Using augmentative biofeedback devices (visual, tactile-kinesthetic and auditory) in speech therapy aimed at reducing hypernasality is also common (McWilliams, Morris & Shelton, 1990; Golding-Kushner, 1995; Kummer, 2014). This is because biofeedback “…provides learners with ongoing physiological performance information that is typically not available to them or has not reached a level of conscious introspection” (Ruscello, 2008b, pp. 298). Theoretically, by making the unconscious and involuntary process of velopharyngeal movement perceptible, it can be manipulated and controlled. Several low technology biofeedback devices are used to improve velopharyngeal closure (e.g., straw, listening tube, stethoscope, mirror, air paddle, or See-Scape device). These devices allow patients to monitor their own sound via auditory or visual cueing (Dworkin, Marunick & Krause, 2004). More advanced technological devices are also commonly used. Delayed auditory feedback sound recording provides auditory and spectral reference points to guide production (Glaze, 2009). The accelerometer provides sound vibration measurements from the exterior surfaces of the nose and neck that can be used to calculate a Horii Oral-Nasal Coupling (HONC) score (Horii, 1980) to guide production (Moon, 2009). The

continuous positive airway pressure (CPAP) device teaches clients to work against adjustable aerodynamic resistance to strengthen the velopharyngeal muscles (Kuehn, 1991; Kuehn et al., 2002). The nasal endoscope guides mislearning by providing direct visualization of the velopharynx (Kummer, 2014) (e.g., patients can be taught the mechanics of maximal velopharyngeal closure and then be given visual feedback during subsequent speech attempts to guide learning) (Hoch et al., 1986). Similarly, using a technique called velopharyngeal biofeedback (VB), a flexible fiberoptic laryngo/velopharyngoscope can also help patients learn how to make changes to their velopharyngeal behaviour by watching the sphincter in real time (Yamaoka et al., 1983; Witzel, Tobe & Salyer, 1988; Shprintzen & Bardach, 1995; Van Lierde et al., 2004). Finally, using the Nasometer, nasalance scores can be projected onto a computer screen in the form of a real-time visual bar display, waveform display or interactive game with adjustable target settings to help guide production (Kummer, 2014).

1.9.1 The issue of volitional control of the velopharyngeal sphincter Overall, management of hypernasality using speech therapy alone is difficult (Yorkston et al., 2001). Shelton and colleagues (1970) reported that while typical speakers could be taught to successfully raise the velum on command, absent of cueing, the same participants were unable to accurately identify their velar position with any consistency. Cassell and Elkadi (1995) indicated that dermal nerve sensors are more established at the front of the oral cavity deteriorating in number as they travel to the back of the mouth. Finally, Hixon, Weismer and Hoit (2008) described velar proprioception (one’s sense of the velum’s position, tension, force and movement) as not accessible to conscious introspection. Kuehn and Moon (1998) inferred that this may be because the velopharyngeal sphincter is concealed inside the body.

Proprioception is critically important for controlling motor function (Riemann & Lephart, 2002). Specific to speech, auditory feedback also plays an important role in control (Guenther, 2006). To date, the main approach to researching feedback mechanisms in speech has been to investigate the role of auditory feedback on modifying different aspects of vocalisation (i.e., intensity, frequency, vowel formants). Albeit two exceptions (de Boer & Bressmann, 2017; de Boer et al., 2019), the speech feedback and compensation literature has not included research examining how auditory feedback influences oral-nasal balance.

Chapter 2 - Control of vocal parameters in speech and song 2.1 Feedback and compensation of intensity, frequency and vowel formants 2.1.1 Feedback and compensation in speech Speech and song exemplify two important ways we communicate. They represent a species- specific call and answer system contingent on two mechanisms: production and perception (Zatorre & Salimpoor, 2013). A phenomenon exemplifying the connection between speech production and perception is the Lombard Effect. The Lombard Effect is a phenomenon wherein speakers involuntarily compensate for ambient noise by elevating their vocal intensity (Lombard, 1911). Lane and Tranel (1971) expanded on this. They described external public and internal private feedback loops. In the public loop, a speaker is focused on communicating with another speaker, so the original speaker will elevate the intensity of their voice in a noisy environment in order to be heard by their conversation partner. In the private loop, a speaker is focused on hearing themselves, so the original speaker will elevate the intensity of their voice in a noisy environment in order to hear themselves better. Lane and Tranel labeled this latter occurrence side-tone amplification effect, wherein speakers adjust to the sound of their own voice (Lane & Tranel, 1971; Chang-Yit, Pick & Siegel, 1975). Foregoing research showing compensation to side-tone has been reported (i.e., speakers increased their loudness when feedback volume was lowered and decreased their loudness when feedback volume was raised) (Siegel & Pick, 1974; Bauer et al., 2006). These studies demonstrate the importance of auditory feedback in regulating vocal intensity.

In addition to intensity, compensation to frequency has also been reported. Elman (1981) described an experiment wherein the fundamental frequency of a group of speakers’ voices was altered upward or downward electronically. Results indicated that speakers produced compensational responses in the opposing direction. Elman’s results were replicated in similar experiments (Kawahara & Williams, 1996; Burnett et al., 1998; Larson et al., 2000; Xu et al., 2004). Compensational articulatory adjustment to perturbed vowel formants has also been documented (Houde & Jordan, 1998; Purcell & Munhall, 2006). Moreover, Purcell and Munhall (2006) found the reactions to be subconscious and automatic. Even when participants were

made aware of the experimental perturbations taking place, their compensatory response was unchanged.

2.1.2 Feedback and compensation in song With respect to control, singing and speech should exhibit similarities. However, there have been only a few research studies investigating feedback and compensation respective to singing. Jones and Keough (2008) investigated how trained and untrained singers utilize auditory feedback to control fundamental frequency. They found that compared to the untrained group, singers compensated less for frequency-altered auditory feedback of the phoneme /ta/, but that they also showed large aftereffects when their feedback was returned to normal. Their results implied that perhaps trained singers rely more heavily on the internal models that guide feedforward control (as opposed to real-time auditory feedback) than amateurs. Their results also implied that pitch representation may be more ingrained in a singer compared to a non- singer because of their training. In a similar study by Scheerer and Jones (2012) using the vowel /a/ as a stimulus, a positive correlation was found between the baseline variability (fluctuations) of an untrained singer’s pitch and the magnitude of their compensation to pitch-altered auditory feedback (i.e. the more variable the baseline, the greater the compensation). Their results demonstrated further that individuals with more stable pitch control may rely more heavily on feedforward processes than those with more variable baseline pitch production who may rely more on feedback.

Using a singing task, Hanrahan (2012) studied the effects of feedback manipulation on eight trained singers’ pitch, intensity, vocal fold contact and resonance control (measured by formant changes). The task involved singing a five note ascending scale followed by a descending major triad sung on the vowel /a/. Hanrahan’s findings indicated that all aspects of voice production were affected by the feedback perturbations, particularly during a procedure that intensified the singer’s formant (e.g., the participants decreased their intensity when the feedback centered around the singer’s formant). However, no statistical analyses of the data were provided. Using professional and non-professional singers, Bottalico, Graetzer and Hunter (2016) explored the effects of loudness perturbation on a singing task (arpeggios sung on /a/ with varied articulation and tempi). They discovered that while both groups increased the intensity of their voices when they heard an increase in accompaniment volume, the professional singers did so to a lesser

degree. Again, this was interpreted as a stronger dependence on internal representations of motor targets in singers versus non-singers.

In summary, several important advances have been made relative to feedback and compensation research in speech and singing. Compensational response to altered auditory feedback of speech has been shown in the areas of frequency (Elman, 1981; Kawahara & Williams, 1996; Burnett et al., 1998; Larson et al., 2000; Xu et al., 2004), intensity (Siegel & Pick, 1973; Bauer et al., 2006), and vowel formants (Houde & Jordan, 1998; Purcell & Munhall, 2006); and compensatory response to altered auditory feedback of singing has been reported in the areas of frequency and intensity with differences indicated between singers and non-singers (Jones & Keough, 2008; Scheerer & Jones, 2012; Bottalico, Graetzer & Hunter, 2016).

2.2 Models of speech motor control The first objective of this thesis was to advance understanding regarding the mechanisms that control oral-nasal balance in speech and singing. Since proprioception in the velopharyngeal sphincter is limited (Hixon, Weismer & Hoit, 2008), it is probable that most speakers and singers rely very largely on auditory feedback to control oral-nasal balance.

Auditory feedback helps to attain and maintain auditory goals based on an amalgamation of the feedback and feedforward control mechanisms of the central nervous system (CNS) (Perkell, 2012). The feedforward mechanism is predictive. It outlines how speech movement is controlled based on past experiences (i.e., learned auditory target regions specific to a desired speech sound). The feedforward mechanism sends the motor commands for speech. The feedback mechanism is reactive. It outlines how speech movement is controlled based on the incoming sensory input the CNS receives, which may indicate deviation from the planned movement (i.e., it monitors production) (Guenther & Vladusich, 2012). If the current state for creating a speech sound is outside the target region for that sound, corrective motor commands are generated. In the domain of speech and language rehabilitation, unexpected sensory feedback sent to the CNS is theorized as able to change speech production superseding the feedforward mechanism’s initial motor commands (Tourville & Guenther, 2011).

Several models of speech motor control exist in the literature (Behrman, 2013). A sampling of models that describe the principles and mechanisms involved in feedback and feedforward control were chosen for description hereafter including: (1) the Task Dynamics Model (TD) (Saltzman & Kelso, 1987; Saltzman & Munhall, 1989); (2) the State Feedback Control Model (SFC) (Houde & Nagarajan, 2011; Hickok, Houde & Rong, 2011); (3) the Task-Dynamics State Feedback Control Model (FACTS) (Ramanarayanan, Van Segbroeck & Narayanan, 2016); and (4) the Directions into Velocities of Articulators (DIVA) model (Guenther, 1995; Perkell et al., 2000; Guenther, 2006; Tourville & Guenther, 2011).

The Task Dynamics Model (TD) (Saltzman & Kelso, 1987; Saltzman & Munhall, 1989) states that desired task states are based upon proprioceptive feedback. However, the information received is not used to generate error or motor commands. Instead the information is used to calculate articulatory gestural scores (i.e., the durations associated with the specific overlapping gestures that make up an utterance, e.g., for the word “bat”, the raised tongue tip gesture and the lip closure gesture), as opposed to articulatory positions (e.g., for the word “bat”, the target position of the tongue tip and lips) (Parrell et al., 2019).

Within the State Feedback Control Model (SFC) (Houde & Nagarajan, 2011; Hickok, Houde & Rong, 2011), the CNS predicts the current dynamic state of what is being controlled (e.g., the tongue) and generates controls based on this information. These predicted internal state estimates (e.g., where is the tongue?) and future state estimates (e.g., where should the tongue be?) undergo correction via comparison between expected feedback and actual feedback.

A combination of the aforementioned models is the Task-Dynamics State Feedback Control Model (FACTS) (Ramanarayanan, Van Segbroeck & Narayanan, 2016). Within this comprehensive model, the SFC’s state prediction and estimation framework is paired with TD’s gestural scoring model.

To date, the most well-known acoustic model of speech motor control is the Directions into Velocities of Articulators (DIVA) model (Guenther, 1995; Perkell et al., 2000; Guenther, 2006; Tourville & Guenther, 2011). The DIVA model is a hybrid system of speech acquisition and production that links feedforward control (i.e., articulatory position) with separate auditory and

somatosensory feedback control loops. The feedforward system sends the motor commands and the feedback systems monitors production. Errors are generated from each control system simultaneously and transformed into corrective motor commands. The articulatory portion of DIVA’s segmental speech sound map corresponds with 10 vocal tract parameters (e.g., lip protrusion, upper and lower lip height, jaw height, tongue height, tongue shape, tongue body position, tongue tip location, larynx height, and glottal opening and pressure). To date however, the map does not include velopharyngeal movement (Tourville & Guenther, 2011).

2.3 Control of oral-nasal balance in speech The control of oral-nasal balance in speech has not been investigated extensively to date. In two studies by de Boer & Bressmann (2017) and de Boer et al. (2019), speakers wore the nasometer and had their nasal signal level made louder or softer, causing them to hear themselves as more nasal or less nasal while they repeated a sentence. In both studies, increased nasal signal level feedback caused speakers to reduce their nasalance scores, while decreased nasal signal level feedback caused speakers to increase their nasalance scores although to a much lesser degree. Srinivas and Bressmann (in press) further reported that knowledge of the nasal perturbation task did not alter the response. Since oral-nasal balance is an important part of singing training (Scotto di Carlo & Autessere, 1987; Alderson, 1993; Sundberg et al., 2007), the main objective of the first project in this thesis was to investigate the effect of altered auditory nasal signal level feedback on oral-nasal balance during a singing task. Since singing teachers have been teaching resonant voice strategies for centuries, a preliminary review of how singing voice pedagogues have evaluated and understood oral-nasal balance constitutes the next chapter.

Chapter 3 - Control of oral-nasal balance in song 3.1 Historical controversy In the early 19th Century, there was a musical movement of Nationalism in Europe wherein composers began to write music representative of their native countries. In Italy, Germany, France and England, Nationalistic methods of singing developed in tandem forming what are now referred to as “The National Schools of Singing”. Across these schools of singing, nasality was both encouraged and discouraged (Miller, 1997).

3.1.1 Schools opposed to nasality The Italian School of Singing generally regarded nasality as a technical fault (Campelo, 2017). Master teacher Pier Francesco Tosi (1653-1732) wrote, “Let the master attend with great care to the voice of the scholar, which… should always come forth neat and clear, without passing through the nose or being choked in the throat (which are the two most horrible defects in a singer, and past all remedy if once grown into a habit)” (Tosi, 1743, pp. 22). Years later, Manuel Garcia II (1805-1906) stated, “The nose is a waste-basket of the brain but not considered for resonance” (Schoen Rene, 1941, pp. 110). From the German School of Singing, Heinrich Ferdinand Mannstein (1806-1872) described three different kinds of sung tones: (1) laryngeal (Kehlton); (2) nasal (Nasenton) and (3) balanced. He referred to the Nasenton tone as “reprehensible” (Mannstein, 1848, pp. 34; Whitener, 2016, pp. 26), encouraging a pinching of the nostrils as corrective. German mezzo-soprano Mathilde Marchesi (1821-1913) also opposed nasality in singing stating that, “…the French language, with its open vowels and nasal sounds, is not only prejudicial, but in direct opposition, to the correct production of the voice” (Marchesi, 1898, pp. 100; Craik, 2011, pp. 28). French mezzo-soprano Blanche Marchesi (1863-1940) was equally averse to nasality in singing stating that humming was almost harmful to a student’s development (Coffin, 1989). Finally, from the English School of Singing, William Huckel (n.d.) encouraged singing with a pure tone that was “…free from any nasal, guttural or dental sound” (Huckel, 1845, pp. 17).

3.1.2 Schools that emphasized nasality While the Italian School generally rejected nasality in singing, it has been suggested that promotion of the Italianate imposto (i.e., impostazione della voce or placement of the voice) may

have inadvertently encouraged some degree of nasality (Miller, 1986; 1996; Hemsley, 1998). From the German School of Singing, Friedrich Schmitt (1812-1884) encouraged nasality, teaching singers to direct their sound through the nasal cavity. Schmitt believed that this gave the voice “…a round, metallic, sonorous sound that would ‘fill a hall'” (Whitener, 2016, pp. 40). He stated it was a ‘happy day’ (Schmitt, 1854, pp. 16) when the student was able to find a nasal tone. Similarly, German soprano Lilli Lehmann (1848-1929) believed that “Singing nasal or toward the nose… cannot be enough studied and utilized. On account of its tonal effect, its noble timbre, it should be amply employed on all kinds of voices. …How little the teachers speak of it is shown by the fact that many singers are quite ignorant of what nasal singing means and when by chance they hear something about it, they are tormented by the idea of ‘singing toward the nose’.” (Lehmann, Aldrich & Willenbücher, 1927, pp. 90-93). Finally, from the French School of Singing, Jean de Reszke (1850-1925) described nasality as an important feature in singing teaching his students to sing “dans le masque” (in the mask, i.e., forward) and “dans le nez” (in the nose) (Johnstone-Douglas, 1989; Bloem-Hubatka, 2012). It is worth noting that Lehmann and de Reszke suffered from voice problems throughout their careers (Bloem-Hubatka, 2012).

In summary, there was widespread disagreement across the schools of singing regarding the presence of nasality in a well-balanced singing tone. Timbral preferences aside, it is likely that there was individualized misunderstanding regarding what sounds nasal and what feels nasal in singing, which contributed greatly to a general lack of consensus regarding what actually defines “nasality” in singing.

3.2 Misunderstood terminology Suggesting that nasality is a misunderstood term in singing has been discussed previously (Titze, 1987; Sataloff & Titze, 1991, 1994; Miller, 1996; Austin, 1997; Callaghan, 2000). Titze (1987) wrote that, “From the singer’s standpoint, there seems to be a perceptual equivalence between nasality and brilliance (ring) in the tone produced… Confusion arises when perception (and particularly auto-perception) is linked to production without careful scrutiny of the acoustic signal” (pp. 34). With respect to this auditory misperception, Jennings and Kuehn (2008) reported that, “In both nasality and vocal ring, the ratio of energy in the high frequency portion of the spectrum (2–4 kHz) to the low frequency portion (0–1 kHz) is increased… Although vocal ring boosts high frequency acoustic energy and nasality tends to flatten the peaks around the first

formant, the ratio of high frequency energy to low frequency energy may play a role in the perception of vocal ring” (pp. 76). Miller (1996) added that, “Nasals tend to have acoustic strength in regions of the harmonic spectrum that are similar in distribution to the harmonic- partial distribution (overtones) found in the well-balanced spectrum of the singing voice” (pp. 29). A spectrographic example of spoken and sung versions of normal and nasalized /i/ is provided in Figure 3.1. Frequency is on the vertical axis and time is on the horizontal axis. Intensity is represented by colour variations (i.e., louder events are indicated by the brighter white colour and quieter events are indicated by the darker black colour). The formants (peaks in the spectra) are represented as coloured horizontal bands. For the spoken vowel /i/, F11 and F2 are far apart (approximately 280-2250Hz), as expected. In the nasalized version, there is a flattening of the spectra in the F1-F2 region. For the sung vowel /i/, F1 and F2 remain far apart and there is a boost in the F3-F5 (2700-4000Hz) region. In the nasalized version, there is a flattening of the spectra in the F1-F2 region and F3-F5 are more spread out. Notably, the spoken nasalized /i/ and the sung normal /i/ also appear to show similar harmonic spectra in the F3-F5 region.

Figure 3.1. Spectrogram for spoken and sung versions of normal /i/ and nasalized /i/

Accordingly, classical singing teachers have used the concept of nasality as a teaching tool for acquiring vocal ring and projection, based on the auditory misperception of what they perceive to

1 In the singing voice science community, a recent trend is to represent the formants F1-F5 as fR1-fR5 (Titze et al., 2015). Since the traditional nomenclature of F1, F2 etc. is still widely used in speech science and linguistics, it will be used throughout the remainder of this text. 28

be nasal (Titze, 1987; Sataloff & Titze, 1991; Austin, 1997; Callaghan, 2000). However, Stark (1999) proposed that from a pedagogical standpoint, it may simply be easier to instruct a singer to be more nasal, or to “get the voice forward” than to ask them to try to boost F2 (pp. 55). However, the concept disregards that during the production of nasalized vowels, the high frequency portion of the spectrum being produced may vary considerably from person to person, so may not be a universally effective teaching strategy.

Concerning vibratory misperception, Miller (1996) described that, “…sympathetic vibration registers subjectively as ‘mask brilliance’ associated with ‘forward placement’ concepts. In actuality, because some of the acoustic properties of nasal phonemes induce sensations similar to well-balanced timbre that is devoid of nasality, confusion often abounds as to when the sound is nasal and when it is not” (pp. 28). According to Titze (1994), since “A vocalist’s sensation of where the vowel is localized (focused) is quite possibly related to the localization of pressure maxima of the standing waves in the vocal tract… It is conceivable that some vocalists rely on these pressure sensations to modify their vowels as needed...” (pp. 167). Respectively, in singing, vocal brilliance is often taught via associative physical sensation that incorporates placing the voice forward by directing the sound toward or around the nose (e.g., toward the bridge of the nose, sinus cavities, cheekbones, palate, teeth, forehead or mask) (Lamperti, 1905; Curtis, 1909; Shakespeare, 1910; Coffin, 1989; Hemsley, 1998). In order to do so, teachers have been known to incorporate phonatory vocal play (e.g., whimpering or sighing) or supportive secondary postures (e.g., smiling, sneering, grimacing, eyebrow raising or nostril flaring) (Miller, 2004). However, while developing a singer’s subjective awareness of localized vibratory feelings near or directly in the nose may be performatively effective, it has also been a major contributor to the confusion that excessive nasality is needed in singing. Overall, since singing “in the mask” can feel nasal, encouraging nasality in singing has been an educationally effective teaching directive to date, albeit not being physiologically accurate.

Many singing teachers also use nasal continuants (e.g., /m/, /n/, /ŋ/) to balance isolated vowel sounds (e.g., /mi-me-ma-mo-mu/; /ŋ-i ŋ-e ŋ-a ŋ-o ŋ-u/) (Miller, 1986). De Reszke (1850-1925) was an advocate of this practice designing vocal exercises based upon a phrase containing a series of nasal continuants that read: “Pendant que l’enfant mange son pain, le chien tremble dans le buisson” (translation: “While the child is eating his bread, the dog is shaking in the

bush”) (Coffin, 1989, pp. 108). In reference to this pedagogical strategy, Titze (1987) explained that, “Teachers may use nasal sounds to trick the student into experiencing some of the key sensations in the facial region, but then [should] quickly point out the alternate (and preferred) ways of obtaining those same sensations. Otherwise, the singer may learn to interpret excessive nasality (twang) as vocal ring. Pinching one's nose during vocalization is one easy way to test for excessive nasality.” (pp. 35-37). Lip trills and tongue trills (i.e., raspberries) have been suggested as suitable alternatives to nasal continuant-loaded exercises in singing (Nix, 1999).

In summation, nasality has been a contentious topic of discussion amongst singing teachers for centuries. Probable causes seem to include: (1) Auditory misperception that nasality and vocal brilliance are equivalent; (2) Vibratory misperception that nasality and vocal brilliance are equivalent; and (3) Overuse of nasal continuants (e.g., /m/, /n/, /ŋ/) to balance the resonance characteristics of vowels, which has exaggerated the importance of nasality in singing. Irrespective of being misunderstood, the concept has proven to be a useful pedagogical tool in singing voice pedagogy to date since nasality can mimic the sensations of a ringing resonant tone. As Austin (2000) conceded: “Some of the most important physical and acoustic principles that could be guiding our in teaching of voice are not widely understood” (pp. 34).

Singing with excessive nasality remains a problem amongst currently developing singers, resulting in poor diction, reduced projection and a lack of timbral beauty (McCoy, 2008, pp. 580). In response, Campelo (2017) recently identified two clarifying terms: (1) nasality and (2) nasal resonance. He described that “Nasality is a term that describes the perception of a vocal timbre, whereas nasal resonance refers to acoustic properties of the sound independent of how it is perceived” (Campelo, 2017, pp. 38). In other words, singing can have nasal resonance, but not necessarily sound nasal. Ultimately, “Suggesting to a student singer that when his or her voice is well-produced, he or she may feel resonance in the vicinity of the nasal passages, is very different from telling him or her to actually let the sound go through the nose.” (Austin, 1997, pp. 220). Therefore, while nasal sounds have been and will likely continue to be used to demonstrate and train resonant voice, it is important that singers and singing teachers comprehend “in no way should resonant voice be confused with nasality” (Titze, 2001, pp. 528).

3.3 Research overview: controlling nasality in singing Looking to the literature, velar control in singing has been discussed by a number of authors. American physician Henry Holbrook Curtis (1856-1920) wrote, “Much has been written on the education of the soft palate in singing, and a great deal of it in our opinion is unnecessary” (Curtis, 1901, pp. 76-77). It has also been a popular speculation that in order to produce a specific timbre, singers rely on highly developed internal motor targets which allow them to control velar height (Scotto di Carlo & Autessere, 1987). Scotto di Carlo (1994) argued that during the production of ringing sounds, the proprioceptive sensitivities of singers are heightened allowing them to sense the movement of the muscles that surround the velum. Alderson (1993, pp. 25) maintained that: “…by training the student’s ears to detect changes in vowel colour, one trains the student to make changes in velar position”. Yet Miller (1996) reported that there is pervasive misunderstanding surrounding a singer’s ability to volitionally control the soft palate and that overall, “there is probably more confusion concerning the role of velopharyngeal closure in singing than about any other acoustic consideration” (pp. 27). The following two sections provide an overview of relevant research examining velopharyngeal control and nasality in singing.

3.3.1 Velopharyngeal control in singing Several studies have investigated the status of the velopharyngeal port during singing, beginning with Russell (1931). Russell took x-rays of famed Italian tenor Enrico Caruso’s (1873-1921) vocal tract while he sang the vowels /i/, /a/, and /u/ (Coffin, 1980). His findings indicated that Caruso’s velopharyngeal port appeared closed for each vowel, which implied that very little nasality was present in his voice. Historical anecdote indicates that Caruso was apparently so outraged by the result that he refused to be identified as the subject in the study. X-rays of Italian baritone Pasquale Amato’s (1878-1942) vocal tract were also taken by Russell (1931). Amato’s velopharyngeal port appeared open on /i/ and /u/, but not on /a/, implying that some nasality was present in his voice. The outcomes were contradictory, possibly due to timbral preference, which could have affected their vocal tract alignment. Nevertheless, the results highlight the early confusion surrounding nasality and singing. Caruso was known for having a ringing quality to his voice, which was always placed (i.e., imposto) in the mask (i.e., forward) in order to localize the focus of his voice (Marafioti, 1922). During the time of Russell’s experiment, mask resonance was often misinterpreted as nasalization (Coffin, 1987). Therefore,

it is probable that Caruso would have suspected that nasality would have been present in his singing voice, and moreover, that he may have felt that a voice that was completely devoid of nasality was not well-produced. Amato’s results complicated things further because they demonstrated that one singer was nasal, and one singer was not. It is probable that the opposing findings from Russell’s experiment may have contributed to the historic controversy surrounding nasality in singing that continued to span across the field of singing voice pedagogy for the next several decades.

Foregoing research on velopharyngeal control in singing continued with Volo and colleagues (1986). Two professional baritones were instructed to sing two-octave arpeggios in a key of their choosing on all five cardinal vowels (i.e., /a e i o u/) in both fast and slow tempi. For both singers, x-ray imaging indicated more complete closure of the velopharyngeal port during the slower sustained singing compared to the florid melismatic (i.e., coloratura) singing. This implied that velopharyngeal closure is task-dynamic subject to singing style. The finding lines up with the work of Troup and colleagues (1989), who stated that “…total velum closure is language dependent, education dependent, style dependent, pitch dependent, and anatomically dependent” (pp. 39).

Specific to pitch, Scotto di Carlo and Autesserre (1987) instructed six professional singers (Fäch not described) representing six different voice classifications to speak and sing the vowels /a i u ɑ̃/ in three pitch ranges (e.g., lower, middle and upper). Velopharyngeal port status was assessed using x-ray imaging and nasal endoscopy. The discussion was restricted to findings from one professional soprano. During production of nasalized /ɑ̃/, the soprano’s velum was more elevated in the high register. This implied that singing at higher pitches may encourage velar elevation. Similarly, Yanagisawa and colleagues (1991) tasked nine professional singers (4M baritones/5F sopranos) and two untrained subjects (2M) to produce vocal sirens as high and as low as possible in several different voice qualities. The larynx, pharyngeal walls and soft palate were assessed using videoendoscopy. Results indicated that when all of the subjects (trained and untrained) sang their highest pitches, laryngeal elevation, lateral pharyngeal wall narrowing, velar elevation and firmer velopharyngeal closure was reported. The outcome supported the work of Scotto di Carlo and Autesserre (1987) suggesting that singing higher pitches may contribute to tighter velopharyngeal closure. Finally, Austin (1997) instructed four classically

trained female singers to speak and sing three sentences, and to sing the vowel series /i e a o u/ at three pitch levels ranging from low to high. The velopharyngeal port area was assessed with a photodetector as described: Two optic fibers were inserted into the nose. The emitter fiber was placed below the velopharyngeal port and the receptor fiber remained in the nasal cavity. As the velopharyngeal port opened, the amount of light transmitted to the receptor fiber increased. Data from the experiment was reported as the duration of time that the velopharyngeal port was open compared to the duration of the sentence stimuli, stated as a percentage. For the group, velopharyngeal opening was greater in speech than in singing, and gradually became smaller as frequency increased. For one subject in particular, findings indicated velopharyngeal opening durations that were significantly different from one another. The duration of opening was 39% for the speaking task, 45% for the low-pitched singing, 50% for the mid-pitched singing and 29% for the high-pitched singing. Their results gave further credence to the notion that velopharyngeal opening tends to decrease as pitch ascends, and that singing at high pitches may induce velopharyngeal closure patterns that are more pronounced than what exists for speech.

Birch and colleagues (2002) assessed the velopharyngeal ports of seventeen classical operatic singers (3 coloratura sopranos, 3 sopranos, 2 mezzo-sopranos, 3 tenors, 2 baritones, 2 bass- baritones, and 2 basses) using nasendoscopy and an airflow mask. All of the singers sang the three words: /panta/, /pinti/, and /puntu/ in different pitch ranges (e.g., A Major triads spanning their whole vocal range) and at different loudness levels (e.g., piano, mezzo forte and forte). Several findings were reported: (1) the opening of the velopharyngeal port was found to increase the level of the singer’s formant; (2) all of the singers used varying degrees of velopharyngeal opening to fine tune the quality of their voices; (3) there was no correlation between velopharyngeal opening and perceived nasality by expert listeners; and (4) operatic tenors exhibited velopharyngeal opening in the passaggio (the transition area between registers). In addition to demonstrating that operatic singers may sing /a/ with a velopharyngeal opening that boosts the singer’s formant, the latter result further implied that nasality may be a useful teaching directive to assist tenors with negotiating their voice breaks. The finding lines up with McCoy (2008) who stated that some male singers use nasality to ease vocal production in the upper passaggio. In a related study by Tanner and colleagues (2005) ten classically trained female singers were tasked to speak and sing the words: /hampa/, /himpi/, and /humpu/ at three intensity levels ranging from soft to loud and at three pitch levels ranging from low to high. Singers were

further instructed to produce the words at a rate of one syllable per second. Nasal airflow, nasal pressure as well as oral airflow were assessed using the PERCI-Speech Aeromechanics Research System, version 3.22 (PERCI-SARS, MicroTronics Corp., Chapel Hill, NC). Velopharyngeal orifice area estimates were generated based on their data. All of the participants demonstrated nasal airflow on non-nasal sounds, and for the group, nasal airflow increased (i.e., 77ml/s compared to 51ml/s) and velopharyngeal port area was calculated as larger (i.e., 3 ㎟ compared to 2 ㎟) during the singing task compared to the speech task. Their results suggest that some trained classical female singers may experience habitual velopharyngeal opening when singing. Whether voice type, style, technique (i.e., forward-focused), or dialect factored into this outcome warrants further investigation. Nonetheless, it is important to note that their findings did not replicate the findings of previous work by Yanagisawa et al. (1991), Scotto di Carlo and Autesserre (1987) or Austin (1997). Finally, using an acoustic model of a baritone vocal tract with a supplemental nasal cavity, Sundberg and colleagues (2007) analysed the effects of velopharyngeal cross-sectional area openings, modelled by differently sized tubes, using computed tomography (CT) scan. Similar to Birch et al. (2002), their results indicated that velopharyngeal opening caused a boost to the level of singer’s formant during production of the vowel /a/. This implied further that velopharyngeal opening may be an effective way to refine vocal brilliance and that professionally trained classical singers may carefully adjust their velopharyngeal port to fine-tune vocal timbre. The finding also lines up with recent work by Gill et al. (2020) who analyzed the acoustic effects of different velopharyngeal openings in eight classically trained singers. The singers were instructed to sing the vowel sequence /i-e-ae-o-u/ in three pitch ranges (e.g., lower, middle and upper passaggio) and in three velopharyngeal opening conditions (i.e., large, narrow and non-existent). Compliance with the three conditions was based on nasal flow, which was monitored and displayed on a computer screen. Findings indicated a significant boost in the 2-4kHz range (i.e., the level of the singer’s formant) for the narrow velopharyngeal opening condition. Similar to Birch et al. (2002) and Sundberg et al. (2007), the results gave further credence to the implication that professional singers may sing with a small velopharyngeal opening in order to enhance vocal brilliance.

From the speech language pathology literature, one further study is worth noting. Bell-Berti and Krakow (1991) studied the velar movement of normal subjects using a Velotrace (Horiguchi & Bell-Berti, 1987). A velotrace is a mechanical device that captures analog data (i.e., data

represented in a physical way) on velar position. The device has an internal and external lever with a push rod between them. The internal lever is placed inside the nasal cavity on the nasal surface of the velum, while the external level rests outside the nose. When the internal lever is raised, the external lever deflects towards the speaker. The levers’ movements can be tracked using a velocity-displacement transducer (i.e., a device that tracks motion) or optoelectronic tracking (i.e., a device that tracks light) (Horiguchi & Bell-Berti, 1987). In Bell-Berti and Krakow’s (1991) study they used the velotrace to track the time-varying velar positions of sustained vowels. Results indicated that when the duration of the vowel /a/ was increased, the accompanying velopharyngeal adjustments were less affected by the coarticulatory influences of the adjacent consonants than during the shorter vocalic utterances. Since singing also involves extended vowel durations, theoretically, engaging in a singing-task could also contribute to less marked coarticulatory movement compared to speech.

3.3.2 Acoustic measurement of nasality in singing Research investigating nasality in singing is sparse, and the results of the studies are somewhat conflicting. In 1938, British voice acoustician E.G. White (1863-1940) stated that resonance occurs in the sinuses, which was highly criticized (White, 1938). The first studies actually examining nasality in singing were conducted by Wooldridge (1956) and Vennard (1964). In Wooldridge’s experiment, six professionally trained singers produced the five cardinal vowels (i.e., /i e a o u/) in two different frequency ranges under normal conditions and when the nasal passages were filled with gauze. Spectral analysis revealed no differences between the normal and abnormal conditions. In Vennard’s experiment, five trained singers produced low-pitched vowels (e.g., /i a u/) under normal conditions and when the nasal passages and nasopharynx were filled with gauze, and maxillary sinuses punctured and filled with 25cc of water. Vennard’s results were similar to Wooldridge’s. No significant perceptual or spectral differences were found between conditions. These studies implied that the nasal cavity was not important in singing.

Bailey (1993) assessed the nasalance scores of nineteen classically-trained male singers singing a portion of Adam Adolphe’s (1803-1856) “O Holy Night” (“…Fall on your knees, Oh hear the angel voices, Oh night divine...”). Nasalance scores for the /a/ in “fall”, the /i/ in “knees”, the /I/ in “hear”, and the /o/ of “voices” were extracted. The most significant finding was that the

vowel /i/ exhibited consistently higher nasalance scores for the word “knees”. The mean score was 24.57%. Scores for the /I/ in “hear” (a similar vowel) were lower (14.24%) (a normative value for /i/ and /I/ is approximately 12% [Lewis, Watterson & Quint, 2000]). The authors proposed that co-articulation of the consonant /n/ may have influenced articulation of the vowel /i/ in “knees” resulting in a higher score. Their findings implied that carryover nasality is important in singing. McIver and Miller (1996) designed a similar experiment. Using thirty trained singers (15M, 15F), they investigated the effect of nasal consonants on oral vowels. Nasalance was the outcome measure. Findings indicated that vowels preceding the nasal continuant /m/ had higher nasalance scores than vowels following the nasal continuant /m/. This implied that the effect of anticipatory nasality is overall greater than carryover nasality in singing.

This led to a series of more recent studies investigating the effect of singing training on nasalance scores. Fowler and Morris (2007) compared the nasalance scores of thirty-six trained singers and thirty-six untrained singers, all female. Each group sang and sustained the vowels /i/, /æ/, /u/, and /a/ for six seconds at three frequency levels, dependent on the vocal range of each singer. No differences were found between groups, indicating that singing training had no significant effect on overall nasal production. For both groups however, nasalance scores were higher during the production of front vowels and when singing lower frequencies. The authors inferred that the increased respiratory and laryngeal effort required to sing at a higher pitch may have induced tighter velopharyngeal closure. They also inferred that the palatopharyngeus muscle may have contracted more in the high register due to the increase in respiratory and laryngeal effort. Jennings and Kuehn (2008) devised a similar study. Forty-six singers (twenty- one amateurs [ten male and 11 female] and twenty-five professionals [eight sopranos, five mezzo-sopranos, six tenors, and six baritones]) produced five-note scalar passages on all five cardinal vowels (i.e., /a e i o u/) in three frequency ranges (e.g., low, middle and high, dependent on each individual’s vocal range) and two dynamic levels (mezzo forte and piano). Specific to frequency for the females, the low range spanned A3-E4 (220-330 Hz), the middle range F4-C5 (350-523 Hz) and the upper range C5-G5 (523-784Hz). Ranges for the men were displaced one octave lower. Findings indicated lower nasalance scores in classically trained singers compared to the amateurs across all pitch ranges and across all of the cardinal vowels, with the exception of /o/. Based on their results, the authors suggested that classical singing training may indirectly

teach students to reduce nasality in their singing, and that increased velopharyngeal closure may be stimulated by the mechanics of singing training.

The first clinical study exploring how singing affects hypernasality in children with cleft palate was by Peter, Abdul Rahman and Pillai (2019). In this study, twenty children (13M, 17F) aged 7-12 years read “the Kampung Passage” (a Malaysian speech assessment tool), and sang a popular Malaysian song filled with both oral and nasal sounds. Results indicated a significant decrease in expert listener’s perception of hypernasality during the singing versus the speech task. They concluded that singing may reduce hypernasality. However, the authors did not report on the phonetic comparability of the singing task compared to the speech task, so the results remain inconclusive.

3.4 Research objective (project 1) Nasality in singing has been wrought with misunderstanding and controversy. Despite decades of research and scholarly debate, the distinction between nasality (i.e., a perceptual phenomenon) and nasalized acoustic resonance was not made clear until recently (Titze, 1991, 2001; Austin, 1997; Campelo, 2017). Research has shown that while some singers may experience a mild degree of velopharyngeal opening during singing (Sundberg et al., 2007; Gill et al., 2020), velopharyngeal closure is primarily tighter in singing compared to speech, particularly during pitch ascent (Yanagisawa et al., 1991; Austin, 1997). Several research studies have also reported that singers may adjust the velopharyngeal port to fine tune oral-nasal balance (Birch et al., 2002; Sundberg et al., 2007), to navigate the passaggio (Birch et al., 2002) or to boost the levels of higher formants (i.e., the singer’s formant) (Sundberg et al., 2007; Gill et al., 2020). It is possible that classical singing training may also inadvertently teach students to reduce nasality in their singing (Jennings & Kuehn, 2008) and that generally speaking, singing may reduce the coarticulatory influence of consonants compared to speech (Bell-Berti & Krakow, 1991).

Previous speech research showed that oral-nasal balance is influenced by altered nasal signal level feedback (de Boer & Bressmann, 2017; de Boer et al., 2019). Investigating the effects of altered nasal signal feedback on oral-nasal balance in singing was the main objective of the first project in this thesis. Knowing that the velopharyngeal sphincter does not have good proprioception (Hixon, Weismer & Hoit, 2008) and knowing what happens when there is a

change to the perception of speech (de Boer & Bressmann, 2017; de Boer et al., 2019), it would be of value to know what happens when there is a change to the perception of song, and to compare the response of singers and non-singers. This would refine present understanding regarding the control of oral-nasal balance.

Chapter 4 - Influence of altered auditory feedback on oral-nasal balance in song

A link to the published paper can be found here: https://www.jvoice.org/article/S0892-1997(18)30207-8/fulltext

4.1 Abstract

Objective: This study explored the role of auditory feedback in the regulation of oral-nasal balance in singing in trained singers and non-singers.

Study design: Experimental repeated measures study.

Methods: Twenty non-singers (10M/10F) and ten female professional singers sang a musical stimulus repeatedly while hearing themselves over headphones. Over the course of the experiment, the nasal level signal in the headphones was increased or decreased so that the participants heard themselves as more or less nasal. Nasalance scores in the different phases of the experiment were quantified using a Nasometer 6450.

Results: A repeated measures ANOVA demonstrated a significant main effect for singing condition F(5, 135) = 3.70, p < .05, and multiple comparison tests demonstrated that the nasalance scores for final baseline and the maximum and minimum nasal feedback conditions were all significantly lower than the first baseline (all comparisons p < 0.05).

Conclusion: There were no differences between the singers and non-singers. All participants had lower nasalance scores in response to both increased and decreased nasal signal level feedback.

Keywords: oral-nasal balance, hypernasality, auditory feedback, singing, nasalance, compensation.

4.2 Introduction The vocal mechanism consists of a source of power (the lungs), an oscillator sound source (the vocal folds), and a filter (the pharynx, oral and nasal cavities) (Fant, 1960). As sound travels through the epilarynx and reaches the pharynx, the velopharyngeal sphincter acts as a valve that enables the speaker to differentiate between oral and nasal sounds. In speech and singing, oral consonants and oral vowels are projected from the oral cavity, nasal sounds (for example /m/, /n/, /ŋ/) are projected from the nasal cavity, and nasalized vowels (for example /ɑ̃/, /ɛ/̃ , /õ/, /œ̃/) are projected from the oral and nasal cavities. The balance of oral and nasal sound is determined by the degree of opening and closing of the velopharyngeal sphincter (Kummer, 2014). The velopharyngeal sphincter consists of a group of muscles attached to the velum and the pharynx. This sphincter is situated between the oral and nasal cavities. A competent velopharyngeal sphincter is fundamental for normal speech, normal singing, and normal swallowing. In order to produce nasal sounds, the velum is lowered, opening the velopharyngeal sphincter. To produce oral sounds, the velum is lifted, closing the port (Hixon, Weismer & Hoit, 2008). For velopharyngeal closure, the superior constrictor, the palatopharyngeus, the levator veli palatini, and the uvular muscle are considered important (Perry, 2011). The mechanisms guiding the control of oral-nasal balance in speech and song are incompletely understood, as proprioception in the velopharyngeal sphincter is not accessible to conscious introspection (Hixon, Weismer & Hoit, 2008). Through conscious introspection and specific training following instruction methods such as Resonant Voice Therapy (Verdolini, 2000), speakers and singers can learn to localize and focus on vibrotactile sensations associated with oral-nasal balance. However, absent such specific training, it is likely that most typical speakers and singers rely in large part on auditory feedback to control oral-nasal balance.

The importance of auditory feedback for different aspects of speech motor control has been demonstrated in previous research. Relative to amplitude, Lombard (1911) described how speakers compensated for ambient noise level increases by unconsciously increasing their own speaking loudness. Lane and Tranel (1971) later termed this the “external” or “public” loop, wherein a speaker is focused on conveying a message to an interlocutor. They contrasted this external loop with a “private” loop” that is created when a speaker’s regular voice-to-ear auditory feedback is substituted with headphone feedback. By altering headphone loudness, Lane and Tranel found that listeners compensate for a volume reduction in feedback intensity of

their own voice by increasing their speaking volume, and vice versa. Subsequent research by Siegel and Pick (1974) demonstrated how a change of auditory feedback could lead to adaptive changes in the speakers’ feedforward motor planning. A recent speech production model guiding adaptation studies is the segmental theory of speech motor control called the DIVA model (Directions into Velocities of Articulators) (Guenther, 1995; Guenther, 1995; Perkell et al., 2000; Guenther, 2006; Tourville & Guenther, 2011). Within this model, speech segments are thought to be coded by the central nervous system as auditory-temporal and somatosensory- temporal goal regions, driven by both feedforward (predictive) and feedback (reactive) mechanisms.

Similar effects of compensation (immediate response to a change in auditory feedback) and adaptation (prospective change of feedforward motor plans based on the altered auditory feedback) have been demonstrated for other aspects of speech motor control. Elman (1981), along with Larson and colleagues (2000) demonstrated that when the speaking pitch of a speaker’s voice was manipulated up or down, speakers produced a compensatory adjustment in the opposing direction. Altered auditory feedback of vowel formants has also been shown to result in compensatory articulatory adjustments to partially offset the effects of the manipulation (Houde & Jordan, 1998; Purcell & Munhall, 2006). These compensation and adaptation reactions are described as unconscious and automatic; even when speakers were informed about the nature of the manipulation, the knowledge did not alter their compensatory reactions (Munhall et al., 2009).

While the control of the singing voice should arguably show commonalities with speech production, there has been very little research about compensation or adaptation in singing. Hanrahan (2012) investigated the effects of frequency-altered feedback on trained singers’ pitch control, vowel formants, vocal fold contact quotient, and intensity in song. His findings indicated that all aspects of voice production studied were affected to some degree by the feedback procedures; compensation being most prevalent during a protocol that intensified the Singers Formant (2500-4000Hz). However, no statistical analysis of the results was provided. Using more rigorous methodology, Jones and Keough (2008) explored the effects of frequency- altered feedback on trained singers and non-singers’ pitch control, and found that trained singers compensated less for frequency-altered auditory feedback, but showed larger aftereffects when

their feedback was returned to normal. These results suggested that trained singers rely more strongly on internal models to update feedforward control than untrained singers do. Moreover, the authors speculated that internal representations of pitch may be more ingrained in singers as a result of their training. In a subsequent study by Scheerer and Jones (2012), a positive correlation was found between baseline variability of an untrained signer’s pitch and their degree of compensation to the altered auditory feedback of pitch, further demonstrating that a more stable internal motor model of fundamental frequency results in a heavier reliance on feedforward control over feedback-driven control. Finally, Bottalico, Graetzer and Hunter (2016) studied the effects of intensity-manipulated auditory feedback on a singing task also using professional and non-professional singers. While both groups increased the intensity of their voices when they heard an increase in accompaniment volume, the professional singers did so to a lesser degree, indicating again a heavier reliance on other internal representations of the motor targets.

An area of speech (and singing) motor control that has not been investigated in much detail to date is the control of oral-nasal balance. Acoustically, nasality adds a low frequency nasal murmur to the spectrum (Dickson, 1962; Hixon, Weismer & Hoit, 2008). Since spectrographic quantification of nasality can be challenging, a convenient and clinically popular method of quantifying the relative contribution of the nasal signal to speech is the calculation of an average nasalance score for a speech sample, using a Nasometer (KayPentax, NJ). The nasalance score is calculated using the formula: nasalance % = nasal/(nasal + oral) x 100 (Fletcher, 1970). Higher nasalance scores indicate more nasality, while lower nasalance scores indicate less nasality (Kummer, 2014). A first study by de Boer & Bressmann (2017) using altered auditory feedback during the production of sentence-level speech demonstrated that increased nasal signal levels led to a compensatory effect on nasalance scores in the opposite direction in speakers of Canadian English. Decreased nasal signal levels however, did not elicit a compensatory reaction of the same magnitude. A second study confirmed the same effects in speakers of Brazilian Portuguese (de Boer et al., 2019).

It would be of interest to better understand the control of oral-nasal balance in singing because it is an important concept in singing instruction. Alderson (1993, pp. 27) states that “…by training the student’s ears to detect changes in vowel colour, one trains the student to make changes in

velar position”. It should be noted that in singing pedagogy, vowel colour references refer to vowel perception based on formant frequency analyses (Delattre et al., 1952). Scotto di Carlo and Autessere (1987) argue that singers rely on acquired internal motor targets to control the height of their velum elevation in singing. Sundberg et al. (2007) state that professionally trained classical singers carefully shape their velopharyngeal port to fine-tune their vocal-timbre. However, there is little experimental research to underpin these beliefs, which are nevertheless quite widely held in singing instruction (Troup et al., 1989; Reid, 1990; Bailey, 1993). Correspondingly, Miller (1996, pp. 27) argues that “there is probably more confusion concerning the role of velopharyngeal closure in singing than about any other acoustic consideration”. Physiologically, it has been observed in numerous studies that velopharyngeal closure is more pronounced in singing compared to speech tasks, especially in the higher range (Yanagisawa et al., 1991; Austin, 1997; Birch et al., 2002; Tanner et al., 2005; Fowler & Morris, 2007). In the only study available on the quantitative acoustic assessment of oral-nasal balance in song, Jennings and Kuehn (2008) described lower nasalance scores in trained classical singers compared to amateur singers. These differences were found in singing tasks in all pitch ranges and on all vowels, with the exception of vowel /o/. Based on their results, the authors speculated that classical singing training may implicitly teach students to reduce nasality in their singing.

The goal of the present study was to assess to what degree altered auditory feedback with increased or reduced nasality in a singing task would lead to changes in nasalance scores. Based on de Boer & Bressmann (2017) and de Boer et al. (2019), it was expected that increased nasal feedback would lead to a compensatory decrease in nasal signal sound level, while decreased nasal feedback would lead to a smaller compensatory increase in nasal signal sound level. Since professionally trained classical singers are said to carefully control their velopharyngeal movement (Sundberg et al., 2007), a comparison of trained singers and non-singers was included in the research design. Based on Jones & Keough (2008), it was expected that professional singers would show a smaller change in nasalance scores in response to the altered auditory feedback. Our hypotheses were the following:

H1 – When the participants heard their nasal feedback increase, they would show a compensatory reaction, indicated by a decrease in nasalance scores H2 – When the participants heard their nasal feedback decrease, they would produce a

compensatory reaction, indicated by an increase in nasalance scores H3 – Professional singers would show a smaller compensatory reaction to the altered auditory feedback than non-singers 4.3 Material and methods 4.3.1 Participants Thirty participants (10 female singers, 10 female non-singers, 10 male non-singers) between the ages of 18-35 years, with a mean age of 24.7 (SD 3.23 years) were recruited for this study. All were native speakers of Canadian English with the accent common to Southern Ontario. The female singers were either students majoring in voice, enrolled in a post-secondary music program or graduates of such a program, currently earning a living as a professional musician. The non-singers had no formal musical training or performance experience. All participants had normal hearing, normal speech, and no history of previous speech therapy, based on self-report. 4.3.2 Stimulus A short song was composed by the last author (see Figure 4.1). The words to the song contained both oral and nasal sounds and were taken from the stimulus sentence used in de Boer & Bressmann (2017): “My hamper was damp so the towels are smelly”. The rhythmic structure was waltz-like in a 3/4 time signature at an allegro moderato tempo of 120 beats per minute. The song was composed in the key of Bb major and the pitch range spanned F3 (174.6 Hz) to D4 (293.7 Hz) so that the song could be sung comfortably by both male and female participants. To cue participants, an introduction consisting of two broken triads in Bb Major and F Major (the dominant fifth) were played before the melody as a prompt. The duration of the song was 25 seconds. Figure 4.1. Singing task.

4.3.3 Participant training The first author sang the stimulus to the participants once before the experiment to orient them to the tune. Participants could also listen to the midi file of the song several times before beginning to sing, and they had the sheet music and lyrics in front of them to follow along. Participants were instructed to sing at a mezzo forte dynamic (medium loud). 4.3.4 Recording procedures Participants wore the Nasometer 6450 headset (Kay Pentax, Montvale, NJ). They sang the melody along to a midi file with a flute sound patch, played back from a computer (Compaq Mini, Hewlett-Packard Enterprise Canada, Mississauga, ON). The continuously looped melody was fed into a mixing console (Xenyx 8, Behringer USA, Bothell, WA) connected to the participants’ headphones (SHL3000RD, Philips Canada, Mississauga, ON). While singing 38 repetitions of the melody, participants received auditory feedback of their voice through headphones. The signals from two additional oral and nasal tie-clip style microphones (ECM- CS3, Sony Canada, Toronto, ON) mounted on the Nasometer’s baffle plate were fed into a multitrack recorder (Tascam DP-008, TEAC America, Montebello, CA). The oral and nasal sound channels were centered in the stereo panorama. Gradual changes to the nasal channel of the Tascam DP-008 multitrack recorder provided more or less nasal sounding feedback. A schematic diagram of the experimental setup is included below (see Figure 4.2).

Figure 4.2. Schematic diagram of equipment for influence of altered auditory feedback on oral-nasal balance in song

The experiment proceeded in the following order: in the first set of 15 recordings of the stimulus, three recordings were made at the 50% baseline 1 setting. Using the sound mixer, feedback from the nasal channel was then increased in 5% increments and six more repetitions of the song were recorded during the ramp-up from 50% to 85%. De Boer and Bressmann (2017) demonstrated that the potentiometers of the particular multi-track recorder used in the experiment reached their maximum at 85% and their minimum at 15%. Turning the potentiometer knob to 100% or 0% had no further effect on the signal volume. Three recordings of the participants were then made at the maximum setting. Then, three final recordings were acquired at the baseline 2 level, with the nasal signal level abruptly turned back to 50% before the first of the three final recordings. At this point, a short rest period was allowed and participants were offered water. After the break, a second set of 15 repetitions of the song were recorded. Three recordings were made for a third baseline at the 50% setting. Using the sound mixer, feedback from the nasal channel was then decreased in 5% decrements and six more repetitions of the song were recorded during a ramp-down from 50% to 15%. Three recordings of participants at the minimum setting were then taken, after which three final recordings were acquired during a fourth baseline. The nasal signal level was abruptly turned back to 50% before the first of the three final recordings.

It was necessary to carry out the recordings in blocks of three repetitions of the song in order to work with the Nasometer software’s recording time limitations (100 seconds). To avoid interruption of the experiment, the participants were instructed to keep singing during a fourth repetition, which was not recorded, while data were saved and the Nasometer recording window was cleared and restarted.

4.3.5 Data analysis Statistical analyses were conducted with Number Cruncher Statistical Software version 8.0 (NCSS, Kaysville, UT). The effect of the nasal feedback levels was analysed with a repeated- measures ANOVA of the mean nasalance scores of averages of three repetitions of the stimulus in six feedback conditions: baseline 1, maximum nasal feedback, baseline 2, baseline 3, minimum nasal feedback, baseline 4. Bonferroni tests were used for further post-hoc testing. 4.4 Results The mean nasalance scores and standard deviations for the groups of speakers in the different feedback conditions can be found in Table 4.1.

Table 4.1. Mean nasalance scores and standard deviations for the groups of speakers in the different feedback conditions (N=30)

Male non-singers Female non-singers Female singers Condition Mean SD Mean SD Mean SD Baseline 1 33.37 5.41 37 6.23 35.13 6.84 Maximum 30.87 5.59 34.6 6.68 33.47 6.15 Baseline 2 31.07 6.11 35.33 6.97 34.17 6.34 Baseline 3 32.27 6.33 35.83 6.91 34.27 6.75 Minimum 30.65 7.8 34.23 7.72 34 6.72 Baseline 4 30.43 7.67 34.77 6.81 34.57 6.55

The overall distribution of the data is shown in the error bar chart below (see Figure 4.3). The figure shows that nasalance was highest in the first baseline. The maximum and minimum feedback conditions resulted in lower mean nasalance scores, and the second and fourth baselines directly following the maximum and minimum feedback conditions, respectively, had numerically lower nasalance scores than both the first and third baselines.

Figure 4.3. Error-bar chart of the average nasalance scores in the different nasal signal level feedback conditions

A repeated measures ANOVA of the nasalance scores was run with group (male non-singers, female non-singers, female singers) as the between subjects variable and feedback condition (baselines 1-4, maximum and minimum nasal feedback) as the within subjects variable. The Geisser-Greenhouse adjusted results showed no significant difference in the mean variance between groups, but a significant main effect for condition F (5, 135) = 3.70, p < .05. Post-hoc Bonferroni multiple comparison tests demonstrated that the nasalance scores for baseline 4 [33.26 (SD 7.08)] and the maximum and minimum nasal feedback conditions [32.98 (SD 6.15), 32.96 (SD 7.36)] were all significantly lower than the first baseline [35.17 (SD 6.16)] (all p < 0.05). 4.5 Discussion This study investigated how singers and non-singers respond to altered auditory feedback of their oral-nasal balance when singing. Nasalance scores were the outcome measure. The first hypothesis stated that when the participants heard their nasal feedback increase, they would show a compensatory reaction, indicated by a decrease in nasalance scores. The results of our statistical analysis revealed that when study participants (singers and non-singers) heard the nasality of their singing increase, they demonstrated lower nasalance scores. This could be taken as evidence to support the first hypothesis and was commensurate with the findings of de Boer & Bressmann (2017).

The second hypothesis stated that when the participants heard their nasal feedback decrease, they would produce a compensatory reaction, indicated by an increase in nasalance scores. Contrary to these expectations, when the participants heard the nasality of their singing decrease, they demonstrated lower nasalance scores. More specifically, nasalance scores were significantly lower than the first baseline, and numerically lower than the third baseline. This finding was different from the observations of de Boer & Bressmann (2017), whose participants showed an inconsistent increase in nasalance scores in response to decreased nasal signal level feedback. It is possible that singers (like speakers) do not perceive decreased nasality (hyponasality) as critically as increased nasality (hypernasality). Based on clinical observations, Shprintzen, Lewin and Croft (1979, pp. 54) argued: “while hyponasal speech is not normal, it is far more desirable than hypernasal speech since the majority of consonant phonemes in the English language have no nasal resonance.” It is possible that this observation applies equally to song.

The third hypothesis was that professional singers would show a smaller compensatory reaction to the altered auditory feedback than non-singers. The results indicated that there were no significant differences in nasalance scores between the singers and the non-singers in their responses to the altered auditory feedback. These results differ from those by Jones and Keough (2008), Scheerer and Jones (2012), and Bottalico, Graetzer and Hunter (2016), all of whom found that singers showed more feed-forward control of their vocal production than non-singers in response to fundamental frequency and amplitude perturbations, demonstrating that the internal models guiding a singer’s control of amplitude and pitch production were more refined than those of untrained singers. However, based on the present study results, the control of oral- nasal balance does not show similar differences related to the level of participants’ singing training in response to altered feedback. It could be argued that oral-nasal balance is an area of mostly non-explicit instruction for singing students, so there may be less opportunity to refine internal models of oral-nasal balance in song. As a result, the trained and untrained singers appeared to have reacted to the altered auditory feedback in comparable patterns.

In the only previous study investigating nasalance scores in singing, Jennings and Kuehn (2008) found lower nasalance scores in classically trained singers compared to amateur singers. However, the singing tasks in their study focused on sustained vowels while the present study involved a complete song with more varied phonetic content. There were also no perturbation conditions in their study design, so the results of the two studies may not be directly comparable.

The observed decrease in nasalance scores in response to decreased nasal feedback was unexpected and intriguing. The altered auditory feedback conditions were always presented in the same order (first increased and then decreased nasal feedback). This was based on an observation in de Boer & Bressmann’s (2017) speech study that the order of presentation had no statistically significant effect on the speakers’ compensatory reactions. However, the auditory manipulations in de Boer & Bressmann (2017) were gradually ramped up and down while the present study abruptly changed the nasal signal levels feedback from the maximum and minimum levels back to baseline. In Figure 3, the numerical differences between the four baseline conditions show that baseline 2 and baseline 3 scores were lower than for the first baseline. This could indicate that participants had not fully recovered from the increased nasal feedback and were showing after-effects. Baseline 4 also showed significantly lower nasalance

scores than the first baseline. Altered auditory feedback can cause speakers to temporarily update their internal motor model for speech production, leading to adaptive changes in feedforward motor planning (Shiller et al., 2009; Patri et al., 2018). Therefore, a possible interpretation for the results was that the participants’ feedforward motor planning had undergone an update during the maximum nasal signal level feedback condition, and that this update was maintained through the remainder of the experiment. In future research, the experiment should be replicated with a reverse order of the feedback conditions, as well as with gradual ramp-downs, as used by de Boer & Bressmann (2017). 4.6 Conclusion The present study investigated the effect of altered auditory feedback on the control of oral-nasal balance in song and found that all participants showed lower nasalance scores in response to both increased and decreased nasal signal level feedback. There were no differences between trained singers and untrained non-singers. Future research should investigate the order of the feedback conditions and the transitions of the auditory feedback back to the baseline in more detail. 4.7 Acknowledgements This research was supported by the Music and Health Science Research Collaboratory (MaHRC) at the University of Toronto. The authors wish to thank all of the participants who took part in the study.

Chapter 5 - Modification of oral-nasal balance in speech and song

5.1 Voice focus in singing: chiaroscuro Voice focus is an important concept in singing instruction with a rich history of discussion related to how to train and equalize resonance (Boone, 1997). In practise, optimal voice focus shares constructs with the finely-tuned chiaroscuro (Kirkpatrick, 2009). “During the eighteenth and nineteenth centuries, the ideal voice quality for classically trained singers was sometimes described as chiaroscuro or ‘bright-dark’ tone... [having] a bright edge as well as a dark or round quality in a complex texture of vocal resonances” (Stark, 1999, pp. 32). As an artistic concept, chiaroscuro was explored in painting, seen in the works of Leonardo Da Vinci (1452-1519), Michelangelo Merisi da Caravaggio (1571-1610), Rembrandt Harmenszoon van Rijn (1606- 1669), and Roger De Piles (1635-1709) (Ganofsky, 2014; Nye, 2014). Achieving a chiaroscuro sound was an ideal that emerged from the Italian School of Singing (Miller, 1997). Master bel canto teachers including Giulio Caccini (1551-1618), Lodovico Zacconi (1555-1627), Giovanni Battista Mancini (1714-1800), Manuel Garcia (1805-1906), as well as Giovanni Battista (1839- 1910) and Francesco (1811-1892) Lamperti prized the chiaroscuro sound describing the vocal quality as more complete and emotionally absorbing (Stark, 1999; Fagnan, 2008).

More recently, chiaroscuro has been described relative to the interaction between the laryngeal source and the resonating vocal tract. Miller (1983) indicated that “…chiaroscuro… designates that basic timbre of the singing voice in which the laryngeal source and the resonating system appear to interact in such a way as to present a spectrum of harmonics perceived by the conditioned listener as that balanced vocal quality to be desired – the quality the singer calls ‘resonant’ ” (pp. 132). Relatedly, Stark (1999) described that “The vocal tract is indeed tractable. Adjustments to the vertical larynx position, the pharynx, the tongue, the jaw and the lips can be coordinated with degrees of glottal closure to produce what Garcia called ‘all the tints of the voice’ ” (pp. 55-56). More recently, Bozeman (2013) specified that chiaroscuro is achieved by means of an open throat (lowered larynx) and a convergent resonator (an articulatory narrowing towards the end of the vocal tract [i.e., the lips]). Overall, chiaroscuro is the highest acoustic endeavour for a Western classical singer affording the voice depth, richness, efficiency, optimized register negotiation, intonation stabilization, the capacity to manage

advanced dynamic contrasts, and the ability to maximize carrying power without adding excessive collision-force pressure to the vocal folds (Davids & LaTour, 2012). Present-day use of the term chiaroscuro would undoubtedly be associated with singing that is clearly operatic in nature (Titze, Maxfield & Walker, 2017).

Based on its acoustic features, the concept of chiaroscuro, the search for a midpoint between chiara and oscuro (light and dark), provide footholds for understanding optimal voice focus (Boone, 1997; Sundberg, 1970). Similarly, in terms of the extreme ends of the two qualities, a voice that lacks balance may be too chiara or too oscuro – too forward or too backward. 5.1.1 Chiara: the forward sound In Western classical singing voice pedagogy, the chiara portion of the sound typically refers to the ‘brightness’, ‘ping’, ‘metal’ or ‘squillo’ (ring) in the sound (Smith, 2007). Historically, excessive chiara production has been attributed to the National French School of Singing, which encouraged singing production that involved an elevated head and tongue position that contributed to both hyoidal and laryngeal elevation (Miller, 1997). In 1837, French tenor Gilbert-Louis Duprez (1806-1896) famously pioneered this kind of production as the first operatic singer to produce a high C5 do di petto or ut de poitrine (from-the-chest) at the Paris Opéra. Duprez’s performance caused a shift in regional operatic vocal technique (Bloch, 2007). “After Duprez, the ut de poitrine became a requirement for the Romantic tenor” (Stark, 1999, pp. 42).

The constricted vocal tract (i.e., high larynx and elevated tongue base) specific to forward voice focus is also popularly associated with non-operatic Contemporary Commercial Music (CCM) singing pedagogy (e.g., belting) (Lawrence, 1979; American Academy of Music Teachers of Singing, 2008). Belting is ‘loud’, ‘brassy’ and ‘bright’ (Estill, 1988; Story, Titze & Hoffman, 1996; Bestebreurtje & Schutte, 2000; Titze et al., 2003) created by a shortening and narrowing of the pharynx with spread lips, elevated tongue base and accompanying laryngeal elevation that reduces the overall supralaryngeal space (Lawrence, 1979; Schutte & Miller, 1988; Schutte & Miller, 1993; Sundberg, Gramming & Lovetri, 1993; Balog, 2005; Titze, 2007; Sundberg, Thalén & Popeil, 2012). Titze, Worley and Story (2011) describe the vocal tract configuration for belt using the term ‘megaphone-shaped’, associated with a divergent, wide open mouth that requires amplification in order to be heard (Lovetri, 2002). Twang, a subcategory of belt, is

‘reedy’, ‘piercing’ and ‘bright’ (Steinhauer & Estill 2008) also created by shortening the vocal tract, but with a co-occurring tightening of the aryepiglottic sphincter (McDonald Klimek, 2008). Twang production corresponds with the same frequency range as the singers formant (3-5kHz) and is often associated with some degree of nasality (McDonald Klimek, 2008; Scearce, 2016). Overall, CCM belt singing is commonly taught using nasal continuants and fronted vowel sounds (e.g., cackling like a witch, quacking like a duck, producing a school-yard taunt-sound [i.e., /njæ njæ, njæ njæ njæ/] or a sheep-sound [i.e., /bæ/]) (Behrman & Haskell, 2013; Welch, Howard & Nix, 2019). While CCM (i.e., forward focused) singing has generally endured a systemic history of association with hyperfunctional voice production, methodological refinement has proven this to be an increasingly less normative opinion (Sundberg, Gramming & Lovetri, 1993; Lovetri, Lesh & Woo, 1999; Stager et al., 2000; Guzman et al., 2015). For example, Titze and Story (1997) describe that in well-produced belt and twang singing, the narrowed epilarynx tends to optimize inertance in the vocal tract allowing the vocal folds to self-sustain vibratory cycles with more efficiency. Perceived advantages of chiara (i.e., forward focused) singing include: attaining tonal ideals, ease of production in higher registers, as well as an improved ability to produce softer dynamics (Hurtado, 2005). 5.1.2 Oscuro: the backward sound In Western classical singing voice pedagogy, the oscuro portion of the sound typically refers to a voice that has ‘darkness, ‘warmth’, ‘bloom’, or ‘depth’ in the sound and is associated with operatic singing (Smith, 2007). Oscuro production (specifically the low-positioned larynx) stems from the German School of Singing (Stockhausen, 1884; Husler & Rodd-Marling, 1976, Miller, 1997), albeit being first discussed by the famed Italian tenor Manuel Garcia in 1840 (Garcia, 1840). Titze, Worley and Story (2011) describe the vocal tract configuration using the term ‘inverted megaphone-shaped’ associated with a convergent mouth opening and produced with a widened oropharynx that does not require amplification in order to be heard (Lovetri, 2002). The widened oropharynx specific to operatic singing was investigated by Ventura and colleagues (2013) who described MRI data showing that trained singers increase the volume of their oral cavity when singing.

Voice focus is predominantly informed by laryngeal position (Bozeman, 2013). Since in Western classical singing voice pedagogy, singing with a high larynx is frequently associated with poor vocal technique (Sundberg & Askenfelt, 1983), most Western classical singers learn to

counteract this vocal behaviour by habitually lowering the larynx when singing. Several authors have described this phenomenon. During pitch ascent, untrained singers were shown to raise the larynx (Shipp & Morissey, 1977; Iwarsson & Sundberg, 1998) as much as 13mm above rest position, while trained singers were shown to lower the larynx as much as 20mm below rest position (Shipp & Izdebski, 1975). While Yanagisawa and colleagues (1991) and Echternach and colleagues (2016) reported dissimilar results for trained singers, it was suggested that their findings may have been due to the fact that their subjects may not have been executing the singing task with good technique (i.e., formant tuning) for the duration of the experiment. With respect to intensity, trained singers tend to lower the larynx as loudness increases (Shipp, 1984; Echternach et al., 2016). Trained singers also tend to sing in a lower laryngeal position than they speak with (Wilbrand, 1992).

However, the optimal manner in which a singer lowers their larynx has been debated across the literature. Nagel (1909) wrote that “The low position of the larynx which is favourable for the full quality of the voice, and which should be anticipated in singing, can be achieved in the same way as by yawning.” (pp. 745; Miller, 1997, pp.86). Stockhausen (1884) relatedly described a laryngeal lowering maneuver involving yawning, depressing the tongue and tucking in the mandible (Miller, 1997). According to McKinney (2005) however, “Singing in the full-yawn or depressed larynx position is an example of artificially maintaining pharyngeal space, which results in a great deal of tension” (pp. 131). Miller’s (1977) assessment of an artificially depressed larynx was that “…freedom of movement cannot take place within the mechanism if a set position is forced upon the musculature” (pp. 90). Correctively, Doscher (1994) wrote that in order to stabilize a lowered larynx, muscular antagonism is required between the laryngeal elevators and depressors in order to avoid static postures. A fundamental shift amongst Western classical singing voice pedagogues occurred thereafter advocating for a ‘comfortably-low’ positioned larynx; achieved by learning to maintain the beginning of; or gesture of a yawn position (McKinney, 2005). This was based on an old dictum from the Italian School of Singing: “chi sa respirare sa cantare” (translation: “He who knows how to breathe, knows how to sing”) (Celloni, in Garcia, 1894, pp. 13; McCoy, 2004; Bozeman, 2013). Italian tenor Enrico Caruso (1873-1921) described a sensation of breathing in the tone (i.e., inhalare la voce, as described by Francesco Lamperti) (Miller, 1986; Stark, 1999), while, tenor Tito Schipa (1889-1965) imagined an egg being slipped into the back of his throat (Mason, 2000). Maintenance of a comfortably-

low positioned larynx has since been recommended across Western classical singing voice pedagogy literature. Overall, perceived advantages of oscuro (i.e., backward focused) singing include: improved supraglottal resonance and carrying power, easier facilitation of register transitions, improved vocal stamina and range, as well as reduced vocal fold contact pressure (Cooper & Cooper, 1977; Shipp, 1987; Doscher, 1994; Sundberg, 1987; Sataloff, 1997; Hurtado, 2005).

Generally, trained singers have been reported as having a refined ability to actively change the shape of their vocal tract (Sundberg et al., 2007) and thereafter, to maintain its position (Shipp & Izdebski, 1975). Investigating how a singing voice pedagogy teaching strategy could inform voice focus maneuvers would be intriguing.

5.2 Voice focus in speech In speech, Boone (1997) described voice focus in terms of a horizontal and vertical cross- sectional model of the vocal tract. He stated that an in-focus balanced voice sounded “…as if it is coming from the middle of the mouth, just above the surface of the tongue” (pp. 71). He further described an out-of-focus voice as either too far forward (raised larynx, protruded tongue, narrowed pharynx), creating a “baby voice” (Boone, 1997, pp. 150) that is very thin, juvenile and bright (sometimes perceived as nasal); or too far backward (lowered larynx, retracted tongue, widened pharynx), creating a “country bumpkin voice” (Boone, 2007, pp. 150) (sometimes perceived as hoarse) that is very dark and throaty (Titze, 1994). In the too-far-forward condition (hereafter denoted forward voice focus), the vocal tract is shortened (see Figure 5.1). Boone considered this posture to be similar to lining up the vocal tract for producing the high front vowel /i/. In the too-far-backward condition (hereafter denoted backward voice focus), the vocal tract is lengthened (see Figure 5.2). Boone considered this posture to be similar to lining up the vocal tract for producing the low back vowel of /a/ (Boone et al., 2010).

Figure 5.1. Forward voice focus diagram.

Figure 5.1. Backward voice focus diagram.

In order to dispel any confusion, it is worth noting that Verdolini (2000) developed a therapeutic intervention called Resonant Voice Therapy (RVT) that is sometimes referred to as “forward focus”. While Boone (1997) discussed voice focus in terms of resonant extremes and how to balance those extremes with respect to the physical state of the vocal tract, Verdolini (2000)’s approach was to manipulate concepts of forward voice focus (i.e., oral and mid-facial vibratory sensations that are considered forward) (Perkins, 1983; Boone, 1997; Verdolini, 2000) in order to teach clients to produce voice in an easy and more resonant way with reduced effort and less

vocal fold contact pressure (Verdolini, 2000; Stemple et al., 2010; Scearce, 2016). In short, the interventions have different goals, so RVT is not completely concomitant with voice focus.

5.3 Formant tuning Like voice focus, formant tuning is another example of manipulating the vocal tract (Sundberg, 1988). Formant tuning means finding the best vocal tract shape to align formants with harmonics in order to boost a given pitch or portion of its acoustic spectrum (Miller, 2008). Formant tuning is a necessary characteristic of both classical and contemporary commercial singing (Miller, 2008). Generally, large pharyngeal resonating cavities tend to vibrate at lower pitches and small pharyngeal resonating cavities tend to vibrate at higher pitches. The first formant is strongly influenced by the shape of the pharynx. The second formant is strongly influenced by the shape of the oral cavity (Bozeman, 2013). Previous research investigating the spectral effects of voice focus adjustments has been explored.

In Western classical singing (C4-C5) (Titze, Maxfield & Walker, 2017), F1 and F2 lower (Sundberg, Gramming & Lovetri, 1993; Stone et al., 2003; Björkner, 2008) and there is a boost to the singer’s formant (a clustering of F3, F4 and F5) (Sundberg 1974; 1990). This “…spectral distance between the vowel-determining formants (F1 and F2) and those formants often referred to as the singer’s formant cluster (F3, F4 and F5)… allows the listener to more clearly distinguish chiaroscuro (bright-dark) timbre in a singer’s voice.” (Titze, Maxfield & Walker, 2017).

In CCM singing (i.e., belt) (G5-C5), F1 and F2 have been documented as higher compared to both conversational speech (Story, Titze & Hoffman, 2001; Sundberg & Thalén, 2010) and Western classical singing (Sundberg, Gramming & Lovetri, 1993; Stone et al., 2003; Björkner, 2008).

Vurma and Ross (2002; 2003) reported similar spectral results (i.e., a boost to F2 in a forward condition, and a decrease in F2 and F3 in a backward condition). Overall, research investigating the spectral effects of laryngeal elevation and depression in non-singers also support these findings (Sundberg & Nordström, 1976). Moreover, Boone and McFarlane’s (1993) critical

review of the yawn-sigh technique (Boone & McFarlane, 1988) which is informed by a low laryngeal position has also indicated the lowering of F2 and F3 respectively.

As stated previously, resonance is informed by two factors: (1) the voice source (i.e., the vibrating vocal folds), and (2) the shaping of the vocal tract (Fant, 1960; Sundberg, 1998). Broadly speaking, vocal fold collision generates harmonics, while vocal tract shaping generates formants. Formant tuning (shaping) informs spectral energy distribution and vowel and consonant definition. It also informs oral-nasal balance. As Sundberg (1998) described, the nose supplements or increases the surface area of the vocal tract adding an additional element to resonance.

Previous research has reported the spectral effects of vocal tract shaping maneuvers (i.e., voice focus) in speech and in singing (Sundberg 1974; 1990; Sundberg & Nordström, 1976; Sundberg, Gramming & Lovetri, 1993; Story, Titze & Hoffman, 2001; Stone et al., 2003; Björkner, 2008; Sundberg & Thalén, 2010; Vurma & Ross, 2002; 2003). It would be intriguing to explore the effects of voice focus maneuvers on oral-nasal balance, particularly because velopharyngeal closure is difficult to control with volition (Kuehn & Moon, 1998; Hixon, Weismer & Hoit, 2008). It is likely that the vocal tract adjustments specific to voice focus will affect air resistance, which may have implications for oral-nasal balance.

5.4 Vocal tract impedance Hagen-Poiseuille’s law (Poiseuille, 1846) describes the flow rate of a fluid as it travels through a cylindrical pipe. The rate is dependent on four factors: the radius and length of the pipe, the viscosity of the fluid and the pressure subjected on the fluid (Sutera & Skalak, 1993). Since air is a gas (i.e., a fluid), this law can be used as an approximation of how differences in the size of the oral cavity affect the flow rate of air. A larger oral cavity (i.e. radius) with all other parameters remaining constant will result in lower resistance. For instance, Cole, Forsyth & Haight (1982) reported reductions in oral resistance related to mouthpiece usage, which increases space in the oral cavity. Accordingly, Nair (2007) stated that during singing, oral aperture expansion increases air flow, which lends itself to reducing air resistance.

With respect to voice focus adjustments, since forward voice focus reduces space in the oral cavity, oral resistance should increase, which would likely force air into the nasal cavity resulting in an increase in nasality. The concept of oral cavity impedance and transpalatal sound transfer into the nasal cavity has been addressed previously (Warren, Duany & Fischer, 1969; Mayo, Warren & Zajac, 1998; Gildersleeve-Neumann & Dalston, 2001; Blanton, Watterson & Lewis, 2015). Since backward voice focus increases oral cavity size, oral resistance should decrease improving oral resonance overall and causing a decrease in nasality. Backward voice focus also adds additional resonating space above the larynx which would likely reduce impedance in the oral cavity even further, raising the prominence of oral resonance overall (Titze, 1987). Enhancing mouth opening to improve oral resonance is also not a new concept (Peterson- Falzone, Hardin-Jones & Karnell, 2001; Boone et al., 2010; Kummer, 2014). Moreover, House and Stevens (1956) proposed that perceptual impressions of oral-nasal coupling may altogether be more informed by acoustic impedances than by velopharyngeal orifice area alone.

Speech science research investigating vowel-induced articulatory settings and the effect on oral- nasal coupling uncovered that sentences loaded with high front vowels (e.g., /i/ and /I/) generate higher nasalance scores than sentences loaded with high back vowels (e.g., /u/ and /ʊ/), low front vowels (/ɛ/ and /æ/), and low back vowels (e.g., /o/ and /a/) (House & Stevens, 1956; Lewis, Watterson & Quint, 2000; Lewis & Watterson, 2003; Awan, Omlor & Watts, 2011; Blanton, Watterson & Lewis, 2015; Ha & Cho, 2015). Correspondingly, using computer modelling, Rong and Kuehn (2012) provided evidence that during synthetic production of the vowel /i/ (to simulate hypernasal speech), lowering the posterior tongue dorsum and advancing the tongue body forward (a common strategy employed to increase velopharyngeal space by expanding the vocal tract) resulted in a reduction in hypernasality, as evidenced via perceptual rating by expert listeners.

5.5 Research overview Examining the role of voice focus on oral-nasal balance in speakers of Canadian English and Brazilian Portuguese was investigated previously by de Boer & Bressmann (2016b) and de Boer et al. (2016). In de Boer & Bressmann (2016b), ultrasound feedback and long-term average spectrum analysis were included in the research design to ensure procedural compliance with the voice focus postures. A condition effect was found for the nasal stimulus (p < 0.01). During the

forward voice focus condition participants’ nasalance scores increased, while during the backward voice focus condition participants’ nasalance scores decreased. In de Boer et al. (2016), ultrasound feedback was again included in the research design. Multiple trainers were also used to refine voice focus training directives and reduce experimenter bias. While a condition effect for stimuli was found, post-hoc tests indicated that across stimuli, the nasalance scores in the backward and normal speaking conditions were found to be lower than the forward voice focus condition, but not from one another (p < 0.05). A trend towards significance was however uncovered at the p < 0.07 level. These studies confirmed the short-term effects of voice focus. They also confirmed that voice focus settings are not harmful to normal participants.

Unlike the previous two studies, a case study by Bressmann et al. (2012) provided report of a speaker with hypernasality that achieved reduced hypernasality by speaking in forward voice focus. While unusual, the authors posited that perhaps the pharyngeal constriction of the forward voice focus posture contributed to improved velopharyngeal closure.

Following up on these studies, Bressmann and colleagues (2017), investigated the role of voice focus on tongue movement. The authors assessed speech rate, as well as cumulative travel distance and posterior measurement angle of the tongue, in three conditions: normal, backward voice focus and forward voice focus. Results indicated no differences in speech rate between conditions. In the forward voice focus condition, there was a condition effect for the posterior tongue (p < 0.05), which moved slower and travelled a smaller cumulative distance compared to the other conditions. In the backward voice focus condition, there was a condition effect for the central tongue (p < 0.05), which moved faster and travelled a larger cumulative distance than the other conditions. The study is beneficial because it showed that voice focus directly affects tongue movement, which would be beneficial for patients with disorders that impact motor control of the tongue (e.g., paralysis). The study also provided further evidence that voice focus maneuvers affect vocal tract shaping further supporting the effectiveness of the procedures.

5.6 Research objectives (projects 2 & 3) Previous speech research has reported that forward voice focus increases nasalance scores and backward voice focus reduces nasalance scores (Rong & Kuehn, 2012; de Boer & Bressmann, 2016b & de Boer et al., 2016). In a single-case study, hypernasal reduction has also been

reported relative to forward voice focus adjustments (Bressmann et al., 2012). Research has also shown that trained singers have a refined ability to adjust their global vocal tract geometry (Sundberg et al., 2007) and to maintain those adjustments (Shipp & Izdebski, 1975). It would therefore be of value to investigate how using singing-voice pedagogy instructional strategies to teach voice focus might affect oral-nasal balance.

Chapter 6 - Influence of voice focus adjustments on oral-nasal balance in speech and song

Santoni, C., de Boer, G., Thaut, M., & Bressmann, T. (2019). Influence of voice focus adjustments on oral-nasal balance in speech and song. Folia Phoniatrica et Logopaedica. doi: 10.1159/000501908

A link to the published paper can be found here: https://www.karger.com/Article/FullText/501908

6.1 Abstract

Objective: This study investigated the effect of training backward and forward voice focus adjustments on oral-nasal balance in speech and singing in typical speakers.

Study design: Experimental repeated measures study.

Methods: Twenty participants (10M/10F) aged 24.25 (SD 3.73) read phonetically balanced, nasal and oral speech stimuli, and sang a song in both forward and backward voice focus conditions. A Nasometer 6450 was used to obtain nasalance scores in the different conditions.

Results: Results indicated that forward voice focus resulted in more nasality (p < 0.01) for the oral stimulus and song. Backward voice focus caused a decrease in nasality (p < 0.01) for the nasal stimulus, the phonetically balanced paragraph and the song. During production of the song, males were more nasal in the forward voice focus condition than females (p = 0.01).

Conclusion: Voice focus can influence oral-nasal balance in normal speakers. More research is needed to investigate whether voice focus adjustments could be helpful in speakers with oral- nasal balance disorders.

Keywords: acoustic measurements, nasalance, nasality, resonance, singing, speech motor coordination, velopharyngeal function, voice training.

6.2 Introduction Oral-nasal balance is an acoustic feature of speech production that allows speakers to differentiate between oral and nasal speech sounds. It is regulated by the opening and closing of the velopharyngeal sphincter. A competent velopharyngeal sphincter is fundamental to the normal execution of both speech and song (Kummer, 2014). The most widely used clinical method for capturing how much sound comes out of the nose during speech is computerized nasometry (Bettens, Wuyts & Van Lierde, 2014). The Nasometer (KayPentax, Montvale, NJ) uses an acoustic baffle plate with separate oral and nasal microphones to measure the relative contribution of the nasal signal to speech. The average nasalance score for a speech sample is calculated using the formula nasalance (%) = nasal/(nasal + oral) x 100 (Fletcher, 1970). Higher nasalance scores indicate more nasality, while lower nasalance scores indicate less nasality (Kummer, 2014).

Nasalance scores in speech are affected by the phonetic content of the speech stimuli. Sentences loaded with high front vowels result in higher nasalance scores than sentences loaded with lower or more posterior vowel sounds (Lewis, Watterson & Quint, 2000; Gildersleeve-Neumann & Dalston, 2001; Awan, Omlor & Watts, 2011; Ha & Cho, 2015). Physiologically, high vowels are associated with higher velar elevation and firmer velopharyngeal closure, while back vowels (e.g. /a/) are produced with a lower velum and diminished velopharyngeal closure force (Moll, 1962; Moon, Kuehn & Huisman, 1994; Kuehn & Moon, 1998).

During singing, the velum has been described as closing more tightly as pitch ascends (Yanagisawa et al., 1991). Since singing involves extended vowel durations, the coarticulatory effects of closing and opening gestures of the velum (Bell-Berti & Krakow, 1991) should be reduced. Presumably, this could result in lower nasalance scores in song than in speech. While Jennings and Kuehn (2008) reported that trained singers produced lower nasalance scores for sung vowels compared to untrained amateurs, Fowler & Morris (2007) did not find a difference.

When the velopharyngeal sphincter is unable to close sufficiently, this results in too much sound energy being emitted through the nasal cavity, causing speech to sound hypernasal (Peterson- Falzone, Hardin-Jones & Karnell, 2010). Hypernasal speech can negatively affect

communication and can result in social stigma (Chapman, Hardin-Jones & Halter, 2003; Watterson et al., 2013; Lee, Gibbon & Spivey, 2017). A frequent congenital craniofacial malformation characterized by the symptom of hypernasal speech is cleft palate (Zajac & Vallino, 2017). It has been reported that there are more dermal nerve sensors at the front of the oral cavity, decreasing in number towards the back of the mouth (Cassell & Elkadi, 1995). Moreover, in most individuals, proprioception in the velopharyngeal sphincter is not accessible to conscious introspection (Hixon, Weismer & Hoit, 2008). Therefore, treating hypernasality using behavioural speech therapy exercises is difficult since most hypernasal speakers cannot simply will their velopharyngeal sphincter to close. This leaves clinicians with limited therapeutic options. Recent research in normal speakers has shown that oral-nasal balance can be modified involuntarily when speakers compensate for altered auditory feedback that makes their speech sound more nasal (de Boer & Bressmann, 2017; de Boer et al., 2019). In a follow-up experiment using a song stimulus (Santoni et al., 2018), trained and untrained singers responded to perturbed nasal feedback similarly suggesting that singers’ internal models for controlling oral-nasal balance may not be as refined as those that have been shown to guide other vocal parameters (e.g. frequency and intensity) (Jones & Keough, 2008; Scheerer & Jones, 2012; Bottalico, Graetzer & Hunter, 2016). To date, it is unknown how much speakers with pathological hypernasality would be able to make comparable involuntary adjustments relative to speech or song.

The present study investigated to what extent oral-nasal balance can be influenced in speech and song using voice focus adjustments. Voice focus describes vocal tract shape and length modifications, which affect the perceptual attributes of voice (Boone, 1997; McCoy, 2004). The optimally balanced voice focus, called chiaroscuro (“light-dark”), is an important concept with roots in singing voice pedagogy (Stark, 1999; Fagnan, 2008). While chiaroscuro describes an ideal balance between the two qualities, the present research study was interested in the extreme ends of the light-dark continuum. In an extreme backward voice focus condition, the vocal tract is maximally lengthened and expanded, which is accomplished by lowering the larynx, widening the pharynx, and retracting the tongue. This results in a dark and throat-centered vocal quality (Story, Titze & Hoffman, 2001; Mainka et al., 2015). In an extreme forward voice focus condition, the vocal tract is maximally shortened and constricted by raising the larynx, narrowing the pharynx, and bringing the tongue forward, which results in a bright vocal quality.

Contemporary Commercial Music pedagogy often emphasizes this more forward singing focus (Lawrence, 1979; Estill, 1988; Schutte & Miller, 1993; Sundberg, Gramming & Lovetri, 1993; Titze, 2007; Sundberg, Thalén & Popeil, 2012; Echternach et al., 2014). Physiologically, voice focus adjustments have been shown to result in a different positioning and organization of movement of the tongue. In forward focus, the posterior tongue travels a smaller cumulative distance and at a slower speed while the central tongue moves a larger cumulative distance and at a higher speed in the backward focus condition (Bressmann et al., 2017).

Using computer modelling, Rong & Kuehn (2012) found that during synthetic production of a hypernasal vowel /i/, lowering the posterior tongue dorsum and advancing the tongue body forward resulted in a reduction in hypernasality, according to perceptual ratings by expert listeners. This dovetails with the observation that high front vowels tend to be associated with higher nasalance scores than low back vowels (Lewis, Watterson & Quint, 2000; Gildersleeve- Neumann & Dalston, 2001; Awan, Omlor & Watts, 2011; Ha & Cho, 2015). Conversely, in a case study by Bressmann et al. (2012), speaking with a forward voice focus resulted in a decrease in hypernasality. Based on these findings, de Boer & Bressmann (2016) investigated the effect of focus adjustments in a group of typical female speakers of Canadian English. The study used a non-nasal sentence from the Zoo Passage (Fletcher, 1976), vowel-loaded sentences from Lewis, Watterson and Quint (2000) and a nasally-loaded sentence from the Nasal Sentences (Fletcher, 1976) as the speech stimuli because these are commonly used in nasality assessment (Zajac & Vallino, 2017). However, the non-nasal stimuli were not ideal because they were at an extreme end of the oral vs. nasal continuum, so there was nowhere to go for the speakers when they adjusted their voice focus. The only condition effects were found for the nasal stimulus, where backward voice focus lowered the nasalance scores as predicted by Rong & Kuehn (2012). De Boer et al. (2016b) repeated the study with typical female speakers of Brazilian Portuguese. This study used phonetically balanced stimuli and found that a forward voice focus increased nasalance scores whereas a backward voice focus reduced nasalance scores.

The purpose of the current experiment was to expand on the previous research (de Boer & Bressmann, 2016b; de Boer et al., 2016). In order to test the technique with clinical populations of speakers with hypernasality, it was necessary to develop and test a set of instructions to be used consistently with speakers. In de Boer & Bressmann (2016b), the participants had the

option of reviewing an online ultrasound image of their tongue while they trained the forward and backward voice focus maneuvers. In de Boer et al. (2016), a team of speech-language pathologists jointly provided the instructions. Both these scenarios are different from the feedback that a clinician would typically be able to provide in a therapy session. The present study used a short set of standardized instructions (see Appendix 6.1), which would allow other clinicians and researchers to replicate the task in the future. Based on previous findings (de Boer & Bressmann, 2016b; de Boer et al., 2016), our hypotheses were the following:

H1 – In response to the forward voice focus adjustment, participants would produce higher nasalance scores. H2 – In response to the backward voice focus adjustment, participants would produce lower nasalance scores.

Since voice focus is a concept that originated in vocal pedagogy (Boone, 1997), the study also included a singing task. The purpose of the singing task was to investigate whether the longer vowel durations in a song would result in a more pronounced effect of the voice focus adjustment on oral-nasal balance compared to the speech tasks. While the previous studies (de Boer & Bressmann, 2016b; de Boer et al., 2016) had been carried out with only female speakers, the present study also included male participants in order to investigate possible gender differences in the effects of the technique. Finally, the effects of the voice focus adjustments on speaking pitch and volume were documented and explored. These aspects of the research were explorative, and no specific hypotheses were formulated about expected outcomes. 6.3 Material and methods 6.3.1 Participants Twenty participants (10M, 10F) between the ages of 18-35 years, with a mean age of 24.25 (SD 3.73) served as the participants for this study. Sample size was based on previous research (de Boer & Bressmann, 2016b; de Boer et al., 2016). All of the subjects were native or quasi-native speakers of Canadian English with a Southern Ontario accent. All had normal hearing, normal speech, and no history of prior speech therapy intervention, based on self-report. 6.3.2 Stimuli The spoken stimuli included a non-nasal sentence from the Zoo Passage (“Look at this book with us, it’s a story about a zoo” [Fletcher, 1976]), a sentence heavily loaded with nasal consonants 68

(“Mama made some lemon jam” [Fletcher, 1976]), and the first paragraph from the Rainbow Passage (Fairbanks, 1960) (see Appendix 6.2). The musical stimulus consisted of a song containing both oral and nasal phonemes, used in previous research (Santoni et al., 2018) (see Figure 6.1). The song was 25 seconds long and composed in a 3/4 time-signature at a tempo of 120 beats per minute. Set in the key of Bb major, the pitch span was F3 to D4. This is in the modal range for both men and women (Boone et al., 2010). While F3 can be low for some females with very high voices, none of the participants appeared to have problems singing the song. It was considered preferable to keep the notes in the modal range because this is also the register used in speech production. Higher notes might have also prompted some participants to access their loft register, which can result in tighter velopharyngeal closure (Yanagisawa, 1991; Austin, 1997).

Figure 6.1. Singing task. (Reproduced from: Santoni, C., de Boer, G., Thaut, M., & Bressmann, T. (2018). Influence of altered auditory feedback on oral-nasal balance in song. Journal of Voice, 34(1), 157.e9-157.e15. doi:10.1016/j.jvoice.2018.06.014)

6.3.3 Participant training Before the experiment, the participants read through all of the speech stimuli aloud. The first author then sang the musical stimulus to the participants to familiarize them with the song. Participants also had the sheet music and lyrics in front of them to follow along, and they had the opportunity to practise the song several times before the experiment began.

The present voice focus training protocol (see Appendix 6.1) was based on previous work (de

Boer & Bressmann, 2016b; de Boer et al., 2016). The protocol was developed and taught by the first author, who is a classically trained singer and singing teacher with over 15 years of professional experience with amateur and professional voices. The researcher provided instructions and modelled extreme forward and backward voice focus for the participants. Participants were oriented to the desired perceptual changes to their voice.

For all of the stimuli, participants were instructed to speak and sing at a comfortable mezzo-forte (medium-loud) dynamic level. Breath and posture were loosely monitored, but were not a focus of the teaching protocol. Participants were instructed to sit up straight so that they could breathe freely and efficiently. 6.3.4 Recording procedures Participants wore the Nasometer 6450 headset (KayPentax, Montvale, NJ). During each repeated nasalance recording, the stimuli were always recorded in the same order: oral sentence, nasal sentence, phonetically balanced paragraph, song. The sentences were recorded three times each. The paragraph and song were recorded once. For the song, participants sang the melody accompanied by a midi file with a flute sound patch, played back from a tablet computer (Stream 7 Tablet, Hewlett-Packard Canada, Mississauga, ON) through one headphone speaker (SHL3000RD, Philips Canada) placed over the participants’ left ear.

All study participants underwent both the backward voice focus and forward voice focus training conditions. Half of the participants (5M, 5F) learned the forward voice focus first and half of the participants (5M, 5F) learned the backward voice focus first, crossing over to the alternate voice focus condition after completion of the first. Nasalance recordings of the participants’ speech and singing were made at four time points: at baseline, with the first changed voice focus, with the second changed voice focus, and at the second baseline, after the experiment. 6.3.5 Data analyses All recordings took place in a soundproof booth using the Nasometer II 6450 (KayPentax, Montvale, NJ). The Nasometer headset has two directional microphones on either side of a baffle plate, which is placed perpendicular to the participant’s face between their upper lip and nose. Calibration of the device took place prior to data collection each day according to the manufacturer’s instructions. The Nasometer calculates nasalance scores that reflect the average of the ratio of the oral to the nasal acoustic signal, computed every 8 milliseconds. To calculate

nasalance, the Nasometer filters the acoustic signal with a 300Hz bandpass filtered around a centre frequency of 500Hz. The acoustic recordings from the Nasometer were exported to .wav files and further analyzed with the Multi-Speech 3700 software (KayPentax, Montvale, NJ). The sampling rate of the exported file was 5,512.5Hz with a signal resolution of 16 bit. Despite the low sampling rate and the radical signal filtering, the sound files could be used to measure intensity (in uncalibrated decibels) as well as fundamental speaking frequency (in Hz).

The Number Cruncher Statistical Software version 8.0 (NCSS, Kaysville, UT) was used to conduct the statistical analyses. The effect of voice focus condition on the speech stimuli was analyzed using repeated-measures analyses of variance (ANOVA). Interaction effects between the between-subjects variables (gender and order of intervention) and the within-subjects variables (speaking conditions) on the speech and song stimuli were also investigated. Where sphericity was violated, Greenhouse-Geisser adjustments were used. Bonferroni tests were used for further post-hoc testing. 6.4 Results The average nasalance scores of the four stimuli (oral sentence, nasal sentence, phonetically balanced paragraph, song) in the four different conditions (baseline 1, forward voice focus, backward voice focus, baseline 2) can be found in Table 6.1.

Table 6.1. Mean nasalance scores and standard deviations in percent for the stimuli in the different speaking conditions.

All speakers (N=20) Stimuli Baseline 1 Forward Backward Baseline 2 Oral Sentence 10.60 (5.46) 16.10 (7.76) 9.98 (7.24) 10.50 (5.72) Nasal Sentence 61.97 (7.27) 60.90 (14.61) 44.20 (15.69) 59.52 (5.83) Balanced Paragraph 32.20 (7.11) 34.45 (8.31) 22.40 (9.62) 30.35 (6.05) Song 36.45 (9.20) 41.45 (12.31) 21.95 (9.88) 33.55 (6.88) Male speakers (N=10) Oral Sentence 9.33 (4.75) 14.90 (6.65) 9.07 (7.62) 8.63 (3.39) Nasal Sentence 59.13 (6.16) 65.07 (7.93) 44.90 (15.23) 57.57 (5.05) Balanced Paragraph 29.30 (5.83) 35.80 (6.07) 21.80 (7.86) 27.90 (3.70) Song 34.40 (9.26) 45.90 (11.61) 20.30 (9.49) 31.50 (6.33) Female speakers (N=10) Oral Sentence 11.87 (6.06) 17.30 (8.94) 10.90 (7.13) 12.37 (7.06) Nasal Sentence 64.80 (7.46) 56.73 (18.69) 43.50 (16.94) 61.47 (61.5) Balanced Paragraph 35.10 (7.36) 33.10 (10.24) 23.00 (11.52) 32.80 (7.08) Song 38.50 (9.13) 37.00 (11.88) 23.60 (10.49) 35.60 (7.11)

The overall distribution of the data is shown in Figure 6.2. Based on visual inspection, for the oral sentence, the forward voice focus condition resulted in a marked increase to the nasalance scores. For the nasal sentence, the backward voice focus condition resulted in a marked decrease to the nasalance scores. For the phonetically balanced paragraph and the song, the backward voice focus condition produced nasalance scores that were lower, and the forward voice focus condition resulted in nasalance scores that were higher than the baselines. An extreme outlier was noted for the nasal stimulus during production of the forward voice focus condition. To ensure that this outlier did not affect the results, the repeated measures ANOVA was run with and without this participants’ data. The results were nearly identical between the two analyses, and the significance levels did not change markedly. Therefore, the results presented below include the data of the outlier.

Figure 6.2. Boxplots of the average nasalance scores for the stimuli in the different speaking conditions (N = 20).

Four repeated-measures analyses of variance (ANOVA) tests were run to compare the four conditions (baseline 1, forward voice focus, backward voice focus, baseline 2) on the nasalance scores for each of the stimuli (oral sentence, nasal sentence, phonetically balanced paragraph, song). The between-subjects variables were gender and order of the intervention (forward- backward versus backward-forward). The within-subjects variable was speaking condition. For the oral sentence, the Greenhouse-Geisser corrected results showed a significant main effect for condition [F(3,48) = 11.57, p < 0.001]. Post-hoc comparisons with Bonferroni tests indicated that the nasalance score for the forward voice focus condition (M 16.1%, SD 7.76%) was significantly higher than baseline 1 (M 10.6%, SD 5.46%), baseline 2 (M 10.5%, SD 5.72%), and the backward voice focus condition (M 9.98%, SD 7.24%); (all comparisons p < 0.05).

For the nasal sentence, the Greenhouse-Geisser corrected results showed a significant main effect for condition [F(3,48) = 19.45, p < 0.001]. Post-hoc Bonferroni tests showed that the nasalance score for the backward voice focus condition (M 44.2%, SD 15.69%) was significantly lower than baseline 1 (M 61.97%, SD 7.27%), baseline 2 (M 59.52%, SD 5.83%), and the forward voice focus condition (M 60.9%, SD 14.61%); (all comparisons p < 0.05).

For the phonetically balanced paragraph, the Greenhouse-Geisser corrected results showed a significant main effect for condition [F(3,48) = 22.52, p < 0.001]. Post-hoc Bonferroni tests demonstrated that the nasalance score for the backward voice focus condition (M 22.4%, SD 9.62%) was significantly lower than baseline 1 (M 32.2%, SD 7.11%), baseline 2 (M 30.35%, SD 6.05%), and the forward voice focus condition (M 34.45%, SD 8.31%); (all comparisons p < 0.05).

For the song, the Greenhouse-Geisser corrected results showed a significant main effect for condition [F(3,48) = 33.16, p < 0.001]. Post-hoc Bonferroni tests indicated that the nasalance score for the backward voice focus condition (M 21.95%, SD 9.88%) was significantly lower than baseline 1 (M 36.45%, SD 9.2%), baseline 2 (M 33.55%, SD 6.88%), and the forward voice focus condition (M 41.45%, SD 12.31%). The nasalance score for the forward voice focus condition was also significantly higher than baseline 2; (all comparisons p < 0.05). Additionally, an interaction effect was found for gender and condition [F(3,48) = 4.93, p = 0.01]. This interaction was explained by the fact that males were found to produce higher nasalance scores than females in the forward voice focus voice condition. However, there were no significant main effects for gender and the order of the interventions in any of the ANOVAs.

The average fundamental frequency measures for the oral sentence, nasal sentence and phonetically balanced paragraph in the different conditions can be found in Table 6.2. Data points for the song are not included because the melody restricted pitch variability. Data are presented separately for male and female participants because male voices are usually lower. The average intensity levels are shown in Table 6.3.

Table 6.2. Mean frequency measurements and standard deviations in hertz for the speech stimuli in the different speaking conditions.

Stimuli Baseline 1 Forward Backward Baseline 2 Male speakers (N=10) Oral Sentence 120.11 (11.40) 158.42 (26.84) 163.92 (34.50) 122.07 (12.76) Nasal Sentence 117.99 (15.36) 155.24 (23.50) 161.50 (36.21) 117.47 (13.35) Balanced Paragraph 120.00 (11.79) 154.29 (24.39) 162.68 (33.01) 120.13 (12.56) Female speakers (N=10) Oral Sentence 182.50 (15.26) 217.38 (18.99) 204.16 (18.62) 189.15 (19.42) Nasal Sentence 181.71 (19.57) 215.36 (17.74) 208.52 (21.08) 185.92 (14.56) Balanced Paragraph 180.91 (13.65) 213.92 (11.24) 207.04 (18.88) 185.82 (10.46)

Table 6.3. Mean intensity measurements and standard deviations in uncalibrated decibels for the stimuli in the different speaking conditions. Results are combined for male and female speakers (N=20).

Stimuli Baseline 1 Forward Backward Baseline 2 Oral Sentence 42.20 (3.18) 44.83 (6.59) 46.74 (5.70) 44.58 (2.72) Nasal Sentence 38.66 (2.92) 40.99 (6.79) 43.62 (6.34) 40.97 (2.65) Balanced Paragraph 49.18 (2.88) 50.75 (7.18) 52.32 (4.55) 50.24 (3.20) Song 56.90 (3.84) 59.87 (4.47) 59.05 (3.65) 58.20 (3.98)

Two sets of four repeated-measures analyses of variance (ANOVA) tests were conducted to analyze the effect of the four conditions (baseline 1, forward voice focus, backward voice focus, baseline 2) on the frequency and intensity measurements for each of the stimuli (oral sentence, nasal sentence, phonetically balanced paragraph, song). The between-subjects variables were gender and intervention order (forward-backward versus backward-forward), and the within- subjects variable was speaking condition. No ANOVA was run on the frequency measurements for the song.

For pitch for the oral sentence, the Greenhouse-Geisser corrected results showed a significant main effect for condition [F(3,48) = 22.14, p < 0.001]. There was also a significant main effect for gender [F(1,16) = 77.11, p < 0.001]. Post-hoc Bonferroni tests indicated that the frequency

measurements in the backward voice focus condition (M 184.04Hz, SD 33.97Hz) were significantly higher than baseline 1 (M 151.30Hz, SD 34.58Hz) and baseline 2 (M 155.61Hz, SD 37.95Hz). Similarly, the frequency measurements in the forward voice focus condition (M 187.90Hz, SD 37.77Hz) were also significantly higher than baseline 1 and baseline 2. Across all conditions, females (M 198.30Hz, SD 22.17Hz) produced higher frequencies than males (M 141.13Hz, SD 30.41Hz); (all comparisons p < 0.05).

For pitch for the nasal sentence, there was a Greenhouse-Geisser corrected significant main effect for condition [F(3,48) = 25.97, p < 0.001]. There was also a significant main effect for gender [F(1,16) = 70.20, p < 0.001]. Post-hoc Bonferroni tests showed that the frequency measurements in the backward voice focus condition (M 185.01Hz, SD 37.60Hz) were significantly higher than baseline 1 (M 149.85Hz, SD 36.90Hz) and baseline 2 (M 151.69Hz, SD 37.66Hz). The frequency measurements in the forward voice focus condition (M 185.30Hz, SD 36.90Hz) were also significantly higher than baseline 1 and baseline 2. Across all conditions, females (M 197.88Hz, SD 22.88Hz) had higher fundamental frequencies than males (M 138.05Hz, SD 30.89Hz); (all comparisons p < 0.05).

For pitch for the phonetically balanced paragraph, there was a Greenhouse-Geisser corrected significant main effect for condition [F(3,48) = 31.15, p < 0.001]. There was also a significant main effect for gender [F(1,16) = 85.10, p < 0.001]. Post-hoc Bonferroni tests revealed that the frequency measurements in the backward voice focus condition (M 184.86Hz, SD 34.68Hz) were significantly higher than baseline 1 (M 150.45Hz, SD 33.62Hz) and baseline 2 (M 152.97Hz, SD 35.53Hz); and in the same manner, the frequency measurements in the forward voice focus condition (M 184.10Hz, SD 35.74Hz) were also significantly higher than baseline 1 and baseline 2, respectively. Across all conditions, females (M 196.92Hz, SD 19.43Hz) produced higher frequencies than males (M 139.27Hz, SD 29.06Hz); (all differences p < 0.05).

For intensity for the oral sentence, there was a significant Greenhouse-Geisser corrected main effect for condition [F(3,48) = 6.03, p = 0.01]. Post-hoc Bonferroni tests indicated that the intensity measurements in the backward voice focus condition (M 46.74dB, SD 5.70dB) were significantly higher than baseline 1 (M 42.20dB, SD 3.18dB) (p < 0.05). An interaction effect was found for order and condition [F(3,48) = 3.36, p = 0.05]. Visual inspection of the interaction

data showed that, in the backward voice focus condition, when the order of intervention was forward-backward, participants produced higher intensity levels than when the order of intervention was backward-forward.

For intensity for the nasal sentence, there was a Greenhouse-Geisser corrected significant main effect for condition [F(3,48) = 6.47, p = 0.01]. Post-hoc Bonferroni tests revealed that the intensity measurements in the backward voice focus condition (M 43.62dB, SD 6.34dB) were significantly higher than baseline 1 (M 38.66dB, SD 2.92dB); (p < 0.05). An interaction effect was also found for order and condition [F(3,48) = 3.86, p = 0.04], which was explained by the fact that, in the backward voice focus condition, when the order of intervention was forward- backward, participants produced higher intensity levels than when the order of intervention was backward-forward.

For intensity for the phonetically balanced paragraph, the effect of condition only approached significance [F(3.48) = 3.42, p = 0.0503]. A Greenhouse-Geisser corrected significant interaction effect was found between order and condition [F(3,48) = 6.82, p < 0.01]. In the backward voice focus condition, when the order of intervention was forward-backward, participants produced higher intensity levels than when the order of intervention was backward- forward.

Finally, for intensity for the song, there was a significant main effect for condition [F(3,48) = 3.27, p = 0.03]. Post-hoc Bonferroni tests indicated that the intensity measurements in the forward voice focus condition (M 59.87dB, SD 4.47dB) were significantly higher than baseline 1 (M 56.90dB, SD 3.84dB); (p < 0.05).

Pearson correlation coefficient tests were calculated to determine if there was a relationship between the measurements. The correlations for the pairs nasalance-frequency and intensity- frequency were very weak (r < 0.1) and statistically non-significant. The correlation between nasalance and intensity was statistically significant (p = 0.01), but the correlation was very weak (r < 0.02). 6.5 Discussion This work explored the role of backward and forward voice focus training adjustments on the

regulation of oral-nasal balance in speech and singing. Oral-nasal balance, measured with nasalance scores, was the main outcome measure. The mean nasalance score is a relatively coarse assessment of oral-nasal balance and may not accurately reflect the effect of voice focus adjustments on individual syllables. However, Watterson, Lewis and Foley-Homan (1999) showed that measures of nasalance require stimuli with at least six syllables to ensure reliability of the measure. The Nasometer software also does not allow the examiner to navigate in the signal and identify segments with any accuracy because only a nasalance contour is displayed for analysis. Since the goal of the study was to demonstrate global effects of voice focus changes on oral-nasal balance in speech and song, the mean nasalance score was an appropriate measure.

The first hypothesis was that participants would change their oral-nasal balance in response to the forward voice focus adjustment, resulting in an increase in nasalance scores. Statistical analyses indicated that forward voice focus with a raised larynx and shortened vocal tract did result in more nasality for the oral stimulus. This is in keeping with previous research by de Boer & Bressmann (2016b) and de Boer et al. (2016).

An unexpected finding in this study was that during production of the song, males responded to the forward voice focus condition with significantly higher nasalance scores than females. Post- puberty, the resting position of the larynx is adjacent to cervical vertebrae C6 in women and C7 in men (Hirano, Kurita & Nakashima, 1983) with vocal tract length (vocal folds to lips) measurements of 14-15.5cm for females and 17-18cm for males (Simpson, 2009). Fitch and Giedd (1999) have reported that the male vocal tract is approximately 12.9mm longer than that of a female. This anatomical difference in vocal tract length may have been a factor because the proportionally longer supraglottic pharyngeal cavity of the males may have allowed for greater displacement of the larynx from its resting position in the forward voice focus condition than was the case for females, resulting in a more significant reaction to the laryngeal adjustment.

One female participant produced markedly lower nasalance scores in the forward voice focus condition during production of the nasal stimulus (6%). Upon further inspection, this participant produced similarly hyponasal scores for the other stimuli in the forward focus condition: Oral stimulus (8%), phonetically balanced paragraph (14%) and song (8%). While the nasalance scores for this individual were different from the rest of the group, the finding mirrors those of a

case study by Bressmann et al. (2012) where a hypernasal speaker also presented with reduced nasalance scores when producing speech with a forward voice focus. The authors speculated that the forward focus led to a narrower nasopharyngeal constriction, resulting in improved velopharyngeal closure. Based on her baseline nasalance scores, the individual in the present study had a competent velopharyngeal sphincter. The results for this speaker in the forward focus speaking condition indicate that the effects of voice focus adjustment may lead to different effects in different individuals. In future research, it would be of interest to use nasopharyngoscopic imaging to visualize the actual vocal tract adjustments that speakers make when speaking with a forward or backward voice focus.

The second hypothesis was that participants would change their oral-nasal balance in response to the backward voice focus adjustment, resulting in a decrease in nasalance scores. The data revealed that backward voice focus with a lowered larynx and lengthened vocal tract resulted in a decrease in nasality for the nasal stimulus, phonetically balanced paragraph and song. These findings correspond with the predictions of the computer-model by Rong and Kuehn (2012) as well as prior research by de Boer & Bressmann (2016b) and de Boer et al. (2016).

Since voice focus is a teaching method used in singing voice pedagogy (Boone, 1997), the study also incorporated a singing task as a means of exploring whether the effects of both the forward and backward voice focus adjustments would be more pronounced for the singing task than for the speech tasks. In singing, vowels tend to be prolonged, reducing co-articulatory effects that have been described for velopharyngeal closing and opening gestures (Bell-Berti & Krakow, 1991). Since the phonetic content of the song differed from the speech stimuli, the results could not be compared statistically. However, visual inspection of the results in Figure 6.2 show that the results for the song followed a pattern similar to that of the phonetically balanced paragraph. In future research, it would be of interest to compare sung and spoken versions of the same stimuli.

Since the previous voice focus adjustment research included solely female speakers (de Boer & Bressmann, 2016b; de Boer et al., 2016), the present study also included male participants in order to explore possible gender differences in the effects of the technique. During production of the song, males responded to the forward voice focus condition with significantly higher

nasalance scores than females. No other statistically significant differences were found between genders.

The effects of the voice focus adjustments on fundamental frequency and intensity were also explored. For fundamental frequency, females produced higher fundamental frequencies, which was expected (Hixon, Weismer & Hoit, 2008; Boone et al., 2010; Stemple, Glaze & Klaben, 2010). Across both genders, in comparison to baseline 1 and baseline 2, pitch increased for all of the stimuli in both the backward and forward voice focus conditions. The increases were greatest during production of the oral sentence, where males demonstrated an average increase of 43.81Hz in the backward voice focus condition compared to baseline 1, and females demonstrated an increase of 34.88Hz in the forward voice focus condition, compared to baseline 1. In connected speech, mean pitch distribution has been reported to be in the range of 30-40Hz (Boone et al., 2010). In the present experiment, females remained within this normative frequency variability range, and males fell just outside of it. Overall, it was surprising that the backward focus resulted in a higher pitch. The first author modelled a yawn-sigh, which lengthens the vocal tract and lowers the larynx, so it could be expected that this would result in a lower pitch. However, the pitch differences were so close to the range of normal variability that the perceptual impact for a listener would have been minimal. Therefore, the overall effect of the voice focus adjustments on pitch appear to have been limited.

For intensity of both the oral and nasal sentences, the backward voice focus condition resulted in significantly higher intensity scores than baseline 1. For the song, the forward voice focus condition resulted in higher intensity values than baseline 1. Across all participants, the numerically largest difference was calculated as 4.96dB. Previous research by Gramming, Sundberg and Akerlund (1991) reported that during soft and loud sustained phonation, vocal loudness fluctuates within a range of 3-5dB. Boone et al. (2010) reported a normal intensity variability range of 10dB for unemotional sentences. Based on these normative values, the present findings fall within the expected normal range. Interestingly, when the order of voice focus instructions was forward first, then backward, intensity scores in the backward voice focus condition were higher across all of the speech stimuli. Albeit, the single largest difference found was 8dB, falling within a normal prosodic range.

Previous research has shown that untrained subjects tend to elevate the larynx when pitch is raised (Shipp & Morissey, 1977; Iwarsson & Sundberg, 1998). For trained singers, both laryngeal lowering (Shipp & Izdebski, 1975), and raising (Yanagisawa et al., 1991; Echternach et al., 2016) have been observed relative to increased pitch. When loudness is increased, the larynx has been shown to lower in trained singers (Shipp, 1984; Echternach et al., 2016). Therefore, the study investigated whether there was a systematic relationship between the nasalance scores and the frequency or intensity measurements. No significant correlation between nasalance and frequency or intensity were found. Relative to intensity, the voice focus adjustments caused small spontaneous intensity changes that were not large enough to affect nasalance scores. Since research has found that higher vocal intensity can be associated with lower nasalance scores (Watterson, York & McFarlane, 1994; Van Lierde et al., 2011) as well as diminished anticipatory and carryover nasal airflow (Zajac, Mayo & Kataoka, 1998; Young et al., 2001), it would be of interest in future research to combine intensity and voice focus changes to investigate their possible cumulative effect.

The present study had a number of limitations. The voice instruction was provided by a single researcher. In future research, it will be important to investigate the reproducibility of the effects of the voice focus interventions on nasalance scores (i.e. backward voice focus results in lower and forward voice focus results in higher nasalance scores) when the protocol is administered by different individuals and with different clinical populations. In the present study, the main indicator for the effectiveness of the voice focus adjustment training protocol was the nasalance score since previous research had already confirmed the effect of extreme voice focus adjustments on long-term average spectra (de Boer & Bressmann, 2016b) and tongue position (de Boer et al., 2016). However, future research investigating the usefulness of voice focus adjustments as a treatment technique in speech therapy should assess the effect of the maneuvers on the acoustic spectrum as well as on listener perceptions. Since the participant group was relatively small, the stimuli were always presented in the same order. Future research should also investigate possible order effects. An additional limitation was that the loudness measures extracted from the Nasometer recording files were in uncalibrated decibels, i.e., decibel measurements from the Nasometer were not further validated with a calibrated sound level meter. However, the Nasometer’s purpose is to measure oral and nasal acoustic signals. The baffle plate ensures that the mouth or nose to microphone distance is stable between speakers,

and the dedicated analog to digital signal conversion hardware ensures that the signal quality is consistent. The purpose of the measurements was to investigate whether voice focus adjustments influenced intensity within speakers, so it was the relative change in intensity rather than the precise measure in calibrated decibels that was of interest. Therefore, the intensity measurements obtained with the Nasometer were sufficient for the purposes of the present study.

In hindsight, it would have been interesting to include a spoken version of the hamper sentence to allow a more direct comparison with the song. However, when the study was designed, there were concerns about the length of the recordings and possible fatiguing effects on the subjects, so it was decided to omit the spoken hamper sentence. Based on the visual analysis of Figure 6.2, the rainbow sentence and the hamper song seemed to behave in a qualitatively similar fashion across the different conditions. Nevertheless, future research should include sung and spoken items with the same phonetic content to allow for a direct comparison. It would also be beneficial to include more detailed acoustic analyses of the effects of the voice focus changes on individual phonetic segments in subsequent experiments. 6.6 Conclusion The study corroborated that forward voice focus leads to higher and backward voice focus to lower nasalance scores in most normal speakers. The differences between male and female participants were small. Future research should investigate the application of the instructional protocol from the present study to speakers with hypernasal oral-nasal balance disorders. 6.7 Acknowledgements This research was supported by the Music and Health Research Collaboratory (MaHRC) at the University of Toronto. The authors wish to thank Miss Emily Quan for her help with the frequency and intensity data measurements. We also thank the participants who took part in the study.

Chapter 7 - Immediate effects of voice focus adjustments on hypernasal speakers’ nasalance scores

Santoni, C., Thaut, M., & Bressmann, T. (2020). Influence of voice focus adjustments on hypernasal speakers’ nasalance scores. International Journal of Pediatric Otorhinolaryngology. doi: 10.1016/j.ijporl.2020.110107

A link to the published paper can be found here: https://www.sciencedirect.com/science/article/abs/pii/S0165587620302500

7.1 Abstract

Objective: To explore the immediate effects of voice focus adjustments on the oral-nasal balance of hypernasal speakers, measured with nasalance scores.

Study design: Experimental crossover repeated measures design.

Methods: Five hypernasal speakers (2M, 3F) aged 5-12 (SD 2.7) learned to speak with extreme forward and backward voice focus. Speakers repeated oral, nasal, and phonetically balanced stimuli. Nasalance scores were collected with the Nasometer 6450.

Results: From the average baseline of 34.27% for the oral stimulus, nasalance increased to 46.07% in forward and decreased to 30.2% in backward focus. From the average baseline of 64.53% for the nasal stimulus, nasalance decreased to 64.13% in forward and decreased to 51.73% in backward focus. From the average baseline of 51.33% for the phonetically balanced stimulus, nasalance increased to 58.87% in forward and decreased to 46.2% in backward focus.

Conclusion: Forward voice focus resulted in higher and backward voice focus resulted in lower nasalance scores during speech for a group of hypernasal speakers. However, there was an exception: One male speaker showed decreased nasalance in forward voice focus. Future research should investigate the longer-term effectiveness of the intervention.

Keywords: oral-nasal balance, hypernasality, nasalance, voice focus, speech language pathology, speech therapy.

7.2 Introduction Oral-nasal balance is an attribute of voice quality describing the division of sound between the oral and nasal cavities. It is primarily controlled by a valving mechanism, i.e., the velopharyngeal sphincter, and normal speech depends on its proper functioning. When there is inadequate separation between the oral and nasal cavities, this results in irregular sound transmission through the nose (Kummer, 2014). Hypernasality occurs when the velopharyngeal sphincter cannot adduct completely. This causes excessive sound to escape into the nasal cavity, resulting in the nasalization of voiced speech sounds. Hypernasal speech can impact intelligibility and negatively affect social interaction (Hunt et al., 2005; Collett & Speltz, 2007; Richman et al., 2012; Stock & Feragen, 2018), sometimes resulting in psychosocial problems such as anxiety, depression and low self-esteem (Lockhart, 2003).

A craniofacial condition often characterized by the symptom of hypernasal speech is cleft palate (Zajac & Vallino, 2017). Even after surgery to repair the cleft, patients may still exhibit velopharyngeal insufficiency (Witt & d’Antonio, 1993; Peterson-Falzone, Hardin-Jones & Karnell, 2001). Other reasons for hypernasal speech include non-cleft related velopharyngeal insufficiency caused by mechanical interference (e.g., short velum, large tonsils or adenoids), or surgical ablations related to cancer. Hypernasality can also be symptomatic of velopharyngeal incompetence caused by a neurogenic speech disorder (e.g. dysarthria or apraxia) (Kummer, 2014).

A commonly used assessment tool for the quantification of oral-nasal balance is the Nasometer 6450 (Pentax Medical, Montvale, NJ). The Nasometer includes a headset with two directional microphones on either side of a metal sound separator plate, which is placed on the speaker’s upper lip halfway between the nose and the mouth. The device provides an acoustic measure of nasality by quantifying the amount of acoustic energy emitted from the nose, calculated as the proportion of nasal to oral sound pressure level (SPL). It is reported as a percentage, based on the formula: nasalance (%) = nasal SPL/(nasal SPL + oral SPL) x 100. The higher the nasalance score, the more nasal sound pressure is present in the speech signal (Fletcher, 1976; Dalston, 1989).

Successful rehabilitation of velopharyngeal movement requires both adequate structure and function. When hypernasality is the result of a structural or neurological deficit, behavioural intervention is generally considered ineffective (Kummer, 2014). It is also difficult to influence velopharyngeal movement with speech therapy because the velopharyngeal sphincter offers only minimal conscious proprioception. This makes it difficult to attain volitional control of its movement (Hixon, Weismer & Hoit, 2008). Treatment with speech therapy to improve velopharyngeal closure for speech is therefore only recommended to a limited number of patients with mild to moderate hypernasality with a physically capable velopharyngeal closure mechanism (Golding-Kushner, 1995). Therapeutic techniques that have been suggested include pitch and loudness modification (Watterson, York & McFarlane, 1994; McHenry, 1997; Stemple, 2000; Stemple, Glaze & Klaben, 2010; Boone et al., 2010), perceptual training (Peterson-Falzone et al., 2006), cul-de-sac therapy with pinched nostrils (Morley, 1970; Kummer, 2014), and articulatory drills (Ruscello, Shuster & Sandwisch, 1991; Golding-Kushner, 1995; Stemple, 2000; Stemple, Glaze & Klaben, 2010; Kummer, 2011).

While the velopharyngeal sphincter is the most important regulator of oral-nasal balance, the position and movement of the articulators in the vocal tract also has some influence. Speech research investigating vowel effects on oral-nasal balance has shown that sentences loaded with high front vowels (e.g. /i/ and /I/) result in higher velar elevation and firmer velopharyngeal closure than sentences loaded with low backed vowels (e.g. /o/ and /a/) (Moll, 1962; Bzoch, 1968; Moon, Kuehn & Huisman, 1994; Kuehn & Moon, 1998). Nasal sound pressure levels are higher for high vowels and front vowels (Hirano, Takeuchi & Hiroto, 1966; Clarke & Mackiewicz-Krassowska, 1977). When there is hypernasality, these vowels are perceived as more nasal (Moore & Sommers, 1973). This is also reflected in higher nasalance scores for high and front vowels (Lewis, Watterson & Quint, 2000; Lewis, Watterson & Blanton, 2008; Awan, Omlor & Watts, 2011; Ha & Cho, 2015). The effect can probably be explained by the relative impedance of the oral and nasal cavities (Warren, Duany & Fischer, 1969; Mayo, Warren & Zajac, 1998). As the airspace in the oral cavity is constricted by the raised tongue, more acoustic energy is emitted from the nose. In a typical speaker, this can happen despite complete velopharyngeal closure because the soft palate has some degree of acoustic transparency (Gildersleeve-Neumann & Dalston, 2001; Blanton, Watterson & Lewis, 2015).

Several authors have suggested that hypernasal speech can be reduced by reducing oral impedance and routing sound through the oral cavity via increased mouth opening, low tongue positioning and widening of the pharynx with a yawn-type maneuver (McDonald & Koepp- Baker, 1951; Shelton, Hahn & Morris, 1968; Coston, 1986; Van Demark & Hardin, 1990; Boone & McFarlane, 1994; Peterson-Falzone, Hardin-Jones & Karnell, 2001; Boone et al., 2010; Kummer, 2014). Based on a computer simulation of the vowel /i/, Rong & Kuehn (2012) further reported that expanding the pharynx and protruding the tongue body should reduce the perception of hypernasality.

These ideas appear similar to the shaping of voice focus in singing pedagogy. The concept of voice focus is rooted in the vocal ideal of chiaroscuro. Chiaroscuro has been taught with regularity since the Bel Canto age and refers to an optimum timbral mixture of light and dark tone. The quality is achieved by singing with a lowered larynx, raised velum, and a released forward tongue (Miller, 1986).

In speech-language pathology, Boone (1997) defined a voice that was “in focus” as coming “…from the middle of the mouth, just above the surface of the tongue” (pp. 71), and further explained that voices that were excessively forward (raised larynx, protruded tongue, narrowed pharynx) or backward (lowered larynx, retracted tongue, widened pharynx), were considered “out of focus”. Excessive forward voice focus sounds bright, thin and juvenile, while excessive backward voice focus sounds dark and throaty (Boone et al., 2010).

Bressmann et al. (2012) described how a forward voice focus resulted in reduced nasalance scores in a hypernasal speaker with a speech bulb prosthesis. This inspired de Boer and Bressmann (2016b) and de Boer et al. (2016) to investigate the effect of forward and backward voice focus adjustments on the oral-nasal balance of typical speakers of Canadian English and Brazilian Portuguese. Santoni et al. (2019) then investigated the effect of extreme voice focus (as forward and as backward as possible) on the oral-nasal balance of normal speakers in speech and singing. All three studies demonstrated that speaking (and singing) in a forward focus increased nasalization, whereas speaking (and singing) in a backward focus reduced nasalization, as demonstrated by nasalance scores. Intriguingly, one participant in Santoni et al. (2019) produced markedly lower nasalance scores when speaking in the forward voice focus condition.

Velopharyngeal closure in speech and swallowing is task-dynamic (Flowers & Morris, 1973; Sphrintzen et al., 1974), and velopharyngeal closure has been reported as tighter in singing compared to speech, particularly during pitch ascent (Yanagisawa et al., 1991; Austin, 1997). Research has also shown that singers adjust velopharyngeal port closure configuration based on timbral preference and singing style (Volo et al., 1986; Gramming et al., 1993; Sundberg et al., 2007). Tanner et al. (2005) demonstrated that some trained sopranos may show small velopharyngeal leaks during singing. On the other hand, classical singers have been shown to produce lower nasalance scores during sustained vowel production compared to amateurs (Jennings & Kuehn, 2008). To date, the only clinical study investigating the effect of a singing task on hypernasality was completed by Peter, Abdul Rahman and Pillai (2019). The authors recruited twenty children who had cleft palate, aged 7-12 years old and had them produce a speech task in Malaysian (the Kampung Passage), and sing a common Malaysian song loaded with oral and nasal sounds. Perceptual assessments using a visual analogue rating scale indicated a significant reduction in hypernasality during the singing versus the speech task. However, it was unclear whether the phonetic content of the speech and singing tasks were comparable.

Since previous studies described the effects of voice focus adjustments in typical speakers (de Boer & Bressmann, 2016b; de Boer et al., 2016; Santoni et al., 2019) as well as in a single hypernasal speaker (Bressmann et al., 2012), the present pilot study investigated the effect of forward and backward voice focus adjustments on nasalance scores in a small group of hypernasal speakers. The study aimed to explore the following hypotheses, based on the research results from typical speakers (de Boer & Bressmann, 2016b; de Boer et al., 2016; Santoni et al., 2019):

H1: Forward voice focus will result in higher nasalance scores.

H2: Backward voice focus will result in lower nasalance scores.

Similar to previous work (Santoni et al., 2019), the study also included a song. This was motivated by the observation by Bell-Berti and Krakow (1991) that longer vowel durations

reduce the co-articulatory impact of adjacent consonants on velopharyngeal closure (i.e., the velum is in a stable closed position for longer). The song was included in order to explore whether the longer vowel durations in a singing task would increase the impact of the voice focus adjustments on oral-nasal balance compared to the speech tasks. 7.3 Material and methods 7.3.1 Participants Study participants were recruited from the Palato-Facial Management Clinic at Five Counties Children’s Centre in Peterborough, Ontario. A referring speech language pathologist circulated the research study recruitment flyer to eligible participants with hypernasal speech (with or without a diagnosis of cleft palate) in the speech clinic’s caseload. The flyer described the study purpose as an investigation of the regulation of oral-nasal balance in speech and singing using voice focus. The administration at the Five Counties Children’s Centre and the Office of Research Ethics at the University of Toronto approved the study.

Seven children (3M, 4F), ranging in age from 5-12 years, with a mean age of 7.86 (SD 2.91), served as the participants for this study. All of the participants were screened for hypernasality and identified as eligible (i.e., their speech was hypernasal) for the study by their speech language pathologist. Participants had normal hearing as reported by the speech-language pathologists who referred them to the study. All of the subjects spoke Canadian English with the accent typical to Southern Ontario.

Individual participant background information can be viewed in Table 7.1. Three speakers had repaired hard and soft cleft palate, and of these, one had an associated diagnosis of Pierre Robin Sequence. Two speakers had 22q11 Deletion Syndrome with no manifest clefts, and two speakers presented with velopharyngeal dysfunction absent of a diagnosis.

Table 7.1. Participant information. Participant Information Speaker Gender Age Speech Observations Medical History 1 F 12 History of mixed nasality, but predominantly No known cleft, query submucous cleft and VPI. mildly hypernasal. 22q11. Dysphonic, hypotonic. 2 F 6 Resonance that ranges from acceptable to mild Repaired hard & soft palate cleft, secondary repair: hypernasal with occasional nasal turbulence. posterior pharyngeal flap. Speech production delay. 3 F 7 Moderate-severe hypernasal with frequent Repaired hard & soft palate cleft, short palate length. nasal turbulence. Pierre Robin Sequence. Speech production delay. 4 M 5 Hypernasal. No known cleft or syndrome, query VPI. Severe motor speech issues (query apraxia). 5 F 6 Acceptable resonance, history of fluctuating to No known cleft. mildly hypernasal. Low facial tone, open mouth posture at rest. Quiet voice, weak projection, high pitch. 22q11. 6 M 12 Acceptable to mild hypernasality with audible Repaired hard and soft palate cleft. nasal turbulence. Delayed articulation/speech sound production. 7 M 7 Hypernasal, which increased after profound No known cleft, query VPI or motor coordination car accident and head trauma. difficulties impacting VP function. Delayed articulation/speech sound production, and motor speech.

7.3.2 Stimuli The stimuli used in this study were similar to what was used in previous research (Santoni et al., 2019). Speech materials included two oral sentences from the Zoo Passage (“Look at this book with us. It’s a story about a zoo”) (Fletcher, 1976), the nasal sentence: “Mama made some lemon jam”) (Fletcher, 1976), and a phonetically balanced sentence from the beginning of the Rainbow Passage (“When the sunlight strikes raindrops in the air, they act like a prism and form a rainbow”) (Fairbanks, 1960). The song entitled the “Hamper Song”; used in previous research (Santoni et al., 2018), was reduced to four bars. The song text contained both oral and nasal phonemes and read “My hamper was damp, so the towels are smelly”. Written in 3/4 time, it was composed in B-flat Major with a pitch range of F3-D4 (i.e. 174.6-293.7Hz). This is an appropriate frequency range for children aged 6-10 years (Sorenson, 1989), which was also considered reasonable for the participants aged 5-12 years in the present study. An instrumental introduction made up of two broken triads was played before the melody designed to cue the beginning of the song.

Documented normative mean scores for the chosen speech stimuli have been reported as 11.25% for the Zoo passage, 59.55% for the nasal Jam sentence, and 31.47% for the Rainbow passage (Kummer, 2005). Based upon prior research on normal participants (Santoni et al., 2019), the mean score for the “Hamper Song” was 33.35%. For an oral speech stimulus like the Zoo Passage, scores <28% indicate no hypernasality, scores between 28-40% indicate a mild degree of hypernasality and scores over 40% denote considerable hypernasality (Dalston, Neiman & Gonzalez-Landa, 1993; Smith & Kuehn, 2007; Kummer, 2014). 7.3.3 Participant training Each of the participants had the opportunity to practice the stimuli prior to the experiment. For the speech stimuli, the 12-year-old children (n=2) read through all of the materials on their own. For the children that were 5-7 years of age (n=5), the first author practiced all stimuli with the children to make sure that the nasalance scores were not affected by reading errors. For the song, the first author sang the tune to all the participants several times to orient them to the melody. Participants were then given the opportunity to rehearse the song with the instrumental accompaniment track several times prior to commencement of the experiment. Participants were asked to produce all stimuli in a mezzo-forte (medium-loud) voice.

The voice focus training protocol was based on previous work (de Boer & Bressmann, 2016b; de Boer et al., 2016) and had been formalized into a set protocol (Santoni et al., 2019). The first author (a trained classical singer and experienced post-secondary studio voice instructor) demonstrated the voice focus maneuvers to each of the participants. During each voice focus posture, participants were encouraged to take note of the associated perceptual and sensory qualities specific to each condition. Once the first author decided that the target voice was produced consistently, recordings of the stimuli were made. If the target voice was lost during the recordings, the participants were asked to stop, and teaching resumed.

Instruction for backward voice focus combined the following facilitating techniques: Opening the mouth wide, yawning, tactile laryngeal awareness, as well as unison production with, and mirroring of, the experimenter. Prior to the procedure, study participants were told: “We will place the voice downward and back, creating a large hollow space for our voice to resonate in”. Participants were asked to open their mouth wide and inhale on the gesture of a yawn in order to orient them to maximal laryngeal lowering. In order to maintain the low-positioned larynx,

maintenance of the inhalatory yawn gesture (as if speaking through a stifled yawn-sigh) using a dark, throaty “operatic” voice permeated the procedure. Participants were encouraged to say /a/ several times to practice and maintain the vocal tract posture.

Instruction for forward voice focus combined the following facilitating techniques: Spreading the lips, swallowing, tactile laryngeal awareness, as well as unison production with, and mirroring of, the experimenter. Prior to the procedure, study participants were told: “We will place the voice forward and up creating a small constricted space for our voice to resonate above our larynx”. Participants were asked to spread their lips wide and swallow in order to orient them to maximal laryngeal elevation. They were encouraged to keep the larynx at the high point of their swallow in order to produce a flat and bright “cartoon” voice. Participants were further encouraged to say /i/ on an ascending glissando repeatedly using the “cartoon” voice to practice and maintain the vocal tract posture.

Four participants (2M, 2F) learned the forward voice focus condition first and the backward voice focus second, while the other three participants (1M, 2F) learned the maneuvers in the opposite order. 7.3.4 Recording procedures Recordings took place in a quiet room, and the Nasometer 6450 (Pentax Medical, Montvale, NJ) was used to record all participants. The nasometer was calibrated once in the morning and once in the afternoon on each recording day. Recordings were saved on a hard disk and measured after the experiment was over. Recordings of the stimuli were always made in the following order: Oral stimulus, nasal stimulus, phonetically balanced stimulus, and song. The speech stimuli were repeated three times and the song was sung only once. In order to effectively execute the singing task, participants sang along to an accompaniment track wherein the melody was played with a flute sound through one headphone speaker (SHL3000RD, Philips Canada, Markham, ON) placed over their left ear. The track was played back from a computer tablet (Stream 7 Tablet, Hewlett-Packard Canada, Mississauga, ON). Nasalance scores were obtained during production of all of the stimuli at four time points: baseline 1, voice focus condition 1, voice focus condition 2, and baseline 2.

7.3.5 Data analysis Statistical analyses were conducted using The Number Cruncher Statistical Software version 8.0 (NCSS, Kaysville, UT). Mean nasalance scores were used to characterize changes to individual participants’ oral-nasal balance. An ad-hoc threshold of +/-10% from a participant’s first baseline score was set as a benchmark for meaningful change in oral-nasal balance. The +/-10% cut-off score was chosen because it exceeded the typical test-retest nasalance score variability reported for normal and hypernasal speakers (Seaver et al., 1991; Litzaw & Dalston, 1992; Watterson, Lewis & Brancamp, 2005; Watterson & Lewis, 2006; de Boer & Bressmann, 2014; Lewis, Watterson & Blanton, 2008).

Inferential statistics were also explored. A repeated-measures analyses of variance (ANOVA) was run to compare the four recording conditions (baseline 1, forward voice focus, backward voice focus and baseline 2) on the nasalance scores of the averages of three repetitions of each of the speech stimuli. Speaking condition and stimuli were the within-subjects variables. Due to the substantial heterogeneity of the study group, no between-subjects variables (such as gender and order of intervention) were explored. Fisher’s LSD Multiple-Comparison tests were utilized for further post-hoc analyses. 7.4 Results For all of the participants, individual scores for all of the stimuli in the four voice focus conditions (baseline 1, forward, backward and baseline 2), as well as differences compared to baseline 1 can be found in the Table 7.2.

Table 7.2. Individual nasalance scores (%) and absolute differences for all of the stimuli in the four voice focus conditions (baseline 1, forward, backward and baseline 2) (N = 7, n = 6). All speakers (N=7, n = 6) Stimuli Baseline 1 Forward Backward Baseline 2 Diff (B1-Fw) Diff (B1-Bw) Diff (B1-B2) Oral Stimulus (N=7) 1 30 42 31 20 12 2 -10 2 13 22 15 12 9 2 -1 3 40 57 42 42 17 2 1 4 35 43 25 23 8 -10 -12 5 27 51 22 25 25 -5 -1 6 18 8 27 7 -11 8 -12 7 40 37 31 30 -3 -9 -10 Nasal Stimulus (N=7) 1 59 66 52 51 8 -7 -7 2 64 58 57 55 -6 -7 -9 3 65 79 67 65 13 2 0 4 65 43 37 45 -22 -28 -20 5 63 62 64 61 -1 1 -2 6 67 75 63 64 8 -4 -3 7 71 71 39 64 0 -32 -7 Phonetically Balanced Stimulus (N=7) 1 53 59 42 41 6 -11 -12 2 43 44 39 48 1 -5 4 3 57 71 60 54 14 3 -4 4 47 48 39 45 1 -8 -2 5 43 51 41 39 8 -2 -5 6 55 47 47 44 -8 -7 -11 7 56 65 49 67 9 -7 11 Song Stimulus (n=6) 1 48 63 38 44 15 -10 -4 2 28 58 17 29 30 -11 1 3 46 58 72 48 12 26 2 4 46 61 41 52 15 -5 6 5 37 55 32 37 18 -5 0 6 44 60 35 42 16 -9 -2

Based on visual inspection of the nasalance data for the oral stimulus, it was found that two of the speakers had nasalance scores in the normal range. The baseline nasalance score for participant 2 was 13%, while the score for participant 6 was 18.33%, well below the cut-off score of 28% suggested by Dalston, Neiman and Gonzalez-Landa (1993). Therefore,

participants 2 and 6 were not included in the descriptive and inferential analysis of the results. However, their data are reported in Table 7.2. Participant 7 did not wish to perform the song, so this stimulus was left out of the descriptive and inferential analyses due to the missing data.

The remaining five participants (participants 1, 3, 4, 5 and 7) (2M/ 3F) ranged in age from 5-12 years, with a mean age of 7.4 (SD 2.7). All of these participants produced baseline nasalance scores between 27-40% during production of the oral stimulus. While participant 5 produced a score of exactly 27%, this subject was deemed appropriate for inclusion in the analysis due to the fact that their score was so close to the hypernasal threshold value of 28% (Dalston, Neiman & Gonzalez-Landa, 1993). This decision was also based upon examination of their elevated baseline scores across the other speech stimuli.

The mean scores for all of the speech stimuli in the four voice focus conditions (baseline 1, forward, backward and baseline 2), as well as differences compared to baseline 1 are reported in Table 7.3 and visualized in Figure 7.1. The song was excluded from the descriptive analyses because of the missing data from participant 7.

From the average baseline of 34.27% (SD 6.1) for the oral stimulus, nasalance scores increased to 46.07% (SD 7.94) in the forward and decreased to 30.2% (SD 7.67) in the backward focus condition (largest individual differences for participant 1: +12%, participant 3 +17% and participant 5: +25% for forward focus, and participant 4: -10% for backward focus).

From the average baseline of 64.53% (SD 4.45) for the nasal stimulus, nasalance scores decreased to 64.13% (SD 13.39) in the forward and decreased to 51.73% (SD 13.67) in the backward focus condition (largest individual differences for participant 3: +13% and participant 4: -22% for forward focus, and participant 4: -28% and participant 7: -32% for backward focus).

From the average baseline of 51.33% (SD 5.99) for the phonetically balanced stimulus, nasalance scores increased to 58.87% (SD 9.54) in the forward and decreased to 46.2% (SD 8.69) in the backward focus condition (largest individual differences for participant 3: +14% for forward focus, and participant 1: -11% for backward focus).

A repeated measures ANOVA was run on the data of the hypernasal patients (participants 1, 3, 4, 5 and 7) to evaluate the effects of the four voice focus conditions (baseline 1, forward, backward and baseline 2) on the mean nasalance scores for the speech stimuli. The song was left out of this analysis due to missing data from participant 7. There were significant main effects for condition [F(3,12) = 9.36, p < 0.01) and stimuli [F(2,8) = 48.29, p < 0.01). Fisher’s LSD Multiple-Comparison post-hoc tests revealed that the nasalance scores for the forward voice focus condition (56.36%) were significantly higher than baseline 1 (50.04%), baseline 2 (44.8%) and the backward voice focus condition (42.71%). In addition, the nasalance scores in the backward voice focus condition also produced nasalance scores that were significantly lower than baseline 1 and the forward voice focus condition (all differences p < 0.05). The three stimuli all differed significantly from each other, with highest nasalance scores for the nasal stimulus, and lowest nasalance scores for the oral stimulus, as expected (all differences p < 0.05).

Table 7.3. Hypernasal group mean nasalance scores (%), standard deviations and absolute differences for the speech stimuli in the four voice focus conditions (baseline 1, forward, backward and baseline 2) (n = 5). Hypernasal speakers (n = 5) Stimuli Baseline 1 Forward Backward Baseline 2 Diff Diff Diff (B1- (B1-Bw) (B1- Fw) B2) Oral Stimulus 34.27 (6.1) 46.07 (7.94) 30.2 (7.67) 27.93 (8.55) 12 -4 -6 Nasal Stimulus 64.53 (4.45) 64.13 (13.38) 51.73 (13.66) 57.4 (8.84) 0 -13 -7 Phonetically Balanced 51.33 (5.99) 58.87 (9.54) 46.2 (8.69) 49.07 (11.35) 8 -5 -2 Stimulus Mean 50.04 (13.82) 56.36 (12.52) 42.71 (13.45) 44.8 (15.65) 6 -7 -5

Figure 7.1. Boxplots of the five hypernasal speakers’ mean nasalance scores (%) for the speech stimuli in the four voice focus conditions (baseline 1, forward, backward and baseline 2) (n = 5).

7.5 Discussion The purpose of this study was to explore the immediate effects of forward and backward voice focus adjustments on the oral-nasal balance of children with hypernasality. The data were obtained from a convenience sample of seven children who were identified as eligible by their speech-language pathologist and who volunteered for the study. However, based on their baseline nasalance scores, participants 2 and 6 had nasalance scores for the oral stimulus within the normal range, indicating that they were not hypernasal. Both participants had repaired cleft palate, which may have led their speech-language pathologists to automatically recommend them for inclusion in the study. They were excluded from the data analysis presented in the results section. Inspection of their data in Table 7.2 demonstrates that their nasalance scores increased

with forward focus and decreased with backward focus, as observed for typical speakers in previous research (de Boer & Bressmann, 2016b; de Boer et al., 2016; Santoni et al., 2019).

The first hypothesis stated that, in response to the forward voice focus adjustment (raised larynx and shortened vocal tract), participants’ nasalance scores would increase. For the combined group of the five hypernasal speakers, forward voice focus resulted in considerable increases in nasalance of more than or equal to 10% during production of the oral stimulus. The results of the ANOVA also confirmed that the forward voice focus condition resulted in a significant increase in mean nasalance compared to baseline 1, the backward voice focus and baseline 2 conditions. Individually, forward voice focus resulted in an increase in nasalance during production of the oral stimulus (participant 1, participant 3, participant 5), nasal stimulus (participant 3), and phonetically balanced stimulus (participant 3). These results are similar to previous findings for typical speakers (de Boer & Bressmann, 2016b; de Boer et al., 2016; Santoni et al., 2019). In contrast, participant 4 had a nasalance score for the nasal sentence that was 22% lower than baseline 1 in response to forward voice focus. In two previously described cases of a female hypernasal speaker (Bressmann et al., 2012) and a female typical speaker (Santoni et al., 2019), lower nasalance scores were documented when speaking with a forward voice focus. It is possible that the forward voice focus condition may have led to a narrower velopharyngeal constriction for these speakers. Since variability of the individual morphology of the velopharyngeal sphincter seems to cause speakers to employ different velopharyngeal closure patterns (Skolnick, 1975; Jordan et al., 2017), the effects of voice focus adjustments may also vary between individuals. Future work should include nasoendoscopic assessment or other imaging of the velopharyngeal sphincter.

The second hypothesis indicated that in response to the backward voice focus adjustment (lowered larynx and lengthened vocal tract), participants’ nasalance scores would decrease. For the combined group of the five hypernasal speakers, backward voice focus resulted in a considerable decrease in nasalance during production of the nasal stimulus (-13% on average). The ANOVA results also revealed that the backward voice focus condition resulted in a decrease in nasalance compared to baseline 1 and the forward voice focus conditions. Individually, backward voice focus resulted in a decrease in nasalance during production of the oral stimulus (participant 4), nasal stimulus (participant 4 and participant 7), and phonetically balanced

stimulus (participant 1). The results support previous findings (de Boer & Bressmann, 2016b; de Boer et al., 2016; Santoni et al., 2019).

Bell-Berti and Krakow (1991) reported that longer vowel durations reduce the co-articulatory impact of adjacent consonants on velopharyngeal closure (i.e., the velum is in a stable raised position for longer). Since singing tasks generally prolong vowels, it was expected that the effects of the voice focus adjustments on the song would be more pronounced. Since there was missing data from participant 7 for the song, this stimulus was not included in the descriptive or inferential data analyses. However, the data in Table 7.2 still allow us to discuss individual participants. In the forward voice focus condition, all participants were found to increase their nasalance by 10% or more, similar to results reported in Santoni et al. (2019). In the backward voice focus condition, only participant 1’s nasalance score dropped by -10%. Participants 4 and 5 had nasalance reductions of only -5%, and, intriguingly, participant 3’s nasalance score increased by +26% in response to backward voice focus. While Peter, Abdul Rahman and Pillai (2019) suggested that singing training could be a helpful behavioural intervention for reducing hypernasality, the small group of participants in the present study appeared to show more marked effects for the speech tasks than for the song.

On a group level, there were no significant differences between the first and second baselines. However, upon inspection of individual data, there were several individuals who had nasalance decreases of 10% or more by the final baseline, compared to the initial baseline. Such decreases were found for the oral stimulus (participant 1, participant 4 and participant 7), the nasal stimulus (participant 4), and the phonetically balanced stimulus (participant 1). On the other hand, a nasalance score increase of +11% during production of the phonetically balanced stimulus was noted for participant 7. In future research, the longer-term carryover effects of voice focus adjustments on oral-nasal balance should be investigated.

Since the present study was a first attempt to investigate the impact of voice focus on speakers with hypernasality, it is not possible to explain conclusively why and how the nasalance scores were affected. A number of different authors have advocated for oral sound redirection accomplished via increased mouth opening as a possible means to decrease hypernasality (McDonald & Koepp-Baker, 1951; Shelton, Hahn & Morris, 1968; Coston, 1986; Van Demark

& Hardin, 1990; Boone & McFarlane, 1994). It could be argued that the effects observed in the present study could be completely explained by the changes to the relative impedance of the oral and nasal cavities achieved by differences in mouth opening between the forward and backward voice focus maneuver (Warren, Duany & Fischer, 1969; Mayo, Warren & Zajac, 1998). However, while there may have been a wider mouth opening in the backward focus condition, recent research by Alighieri and colleagues (Alighieri et al., 2020) found that nasalance scores decreased with increased mouth opening in normal subjects, but not in hypernasal speakers with cleft palate. The results for participant 4, who achieved lower nasalance scores in forward focus, are also difficult to explain as a result of reduced oral impedance. In future research, imaging techniques such as nasal endoscopy should be used to investigate if and how voice focus adjustments affect the configuration and movement of the velopharyngeal sphincter during speech.

Based on the results of the present study, it appears that voice focus adjustments may be helpful to reduce hypernasality for select patients. However, while the results of the present research are encouraging, it should be noted that speech therapy alone cannot remediate an anatomically insufficient velopharyngeal sphincter (Kummer, 2014). Voice focus adjustments as a therapeutic tool to address hypernasal speech probably have the greatest chance of success in patients with a physiologically adequate velopharyngeal mechanism. Although voice focus adjustments may have potential as a facilitative technique, extreme backward or forward voice foci are not socially desirable long-term speech qualities. After voice focus adjustments have been used to reduce hypernasality, the clinician would have to gradually phase out the extreme voice focus while maintaining the improved oral-nasal balance. If and how this can be accomplished should be investigated in further research.

The study had a number of limitations. The sample size was small. The convenience sampling approach resulted in a heterogeneous study group. Since this was a first evaluation of the effects of voice focus adjustments on speakers with hypernasality, it seemed reasonable to try it with a small number of participants. Participants 2 and 6 had to be excluded from the results because their nasalance scores for the oral stimulus were in the normal range, and participant 7’s unwillingness to sing limited the statistical analysis of the song data. The effectiveness of the technique was only assessed with nasalance scores. Nasalance was considered the most

100

reasonable measure for the purposes of the present study even though reported correlations between nasalance scores and nasality judgements have been variable (Dalston, Warren & Dalston, 1991; Dalston, Neiman & Gonzalez-Landa, 1993; Hardin et al., 1992; Kuening et al., 2002; Brunnegård, Lohmander & Van Doorn, 2012). Since listener assessments of hypernasal speech can be marred by low listener agreement (Kuening et al., 2002), the added difficulty of assessing nasality in the presence of extreme voice focus adjustments would have added complexity to the listeners’ task. Future research should include additional auditory-perceptual evaluations. 7.6 Conclusion For the combined group of the five hypernasal speakers, forward voice focus resulted in higher, and backward focus resulted in lower nasalance scores. However, there was an exception: One male participant produced lower nasalance scores in the forward voice focus condition during production of the nasal stimulus. More research is needed to further investigate the potential of voice focus adjustments as a possible therapy technique for speakers with hypernasality. Future research should include nasopharyngoscopic or videofluoroscopic imaging to visualize the effect of voice focus adjustments on the velopharynx as well as auditory-perceptual assessments. The longer-term effectiveness of the intervention as a therapy technique should also be investigated. 7.7 Acknowledgements This research was supported by the Music and Health Research Collaboratory (MaHRC) at the University of Toronto. The authors wish to thank the clinicians and staff at Five Counties Children’s Centre for their support and collaboration, especially Leslie Piekarski and Marika Beaumont. We also offer our thanks to all of the children who took part in this study.

101

Chapter 8 - Conclusion 8.1 Oral-nasal balance control The research that makes up this dissertation was designed to further add to our understanding of the control and modification of oral-nasal balance in speech and in singing.

Chapter 1 explored the historical origins and evolution of the concept of vocal resonance. It also defined the term oral-nasal balance as a specific aspect of resonance and provided an anatomical description of the anatomical structure that controls oral-nasal balance, i.e., the velopharyngeal sphincter. Assessment and treatment of velopharyngeal dysfunction was reviewed. The chapter also highlighted that conscious control of oral-nasal balance is difficult because of limited proprioception in the velopharyngeal sphincter (Hixon, Weismer & Hoit, 2008).

Chapter 2 briefly described several models of speech motor control, and provided a review of literature about compensation responses to altered auditory feedback in speech and singing in the areas of frequency (Elman, 1981; Larson et al., 2000; Jones & Keough, 2008; Scheerer & Jones, 2012), amplitude (Lombard, 1911; Lane & Tranel, 1971; Bottalico, Graetzer & Hunter, 2016), and vowel formants (Houde & Jordan, 1998; Purcell & Munhall, 2006). The chapter also described recent research exploring the effect of altered auditory nasal signal level feedback on the regulation of oral-nasal balance in speech (de Boer & Bressmann, 2017; de Boer et al., 2019).

Chapter 3 provided an overview of the concept of nasality in singing voice pedagogy and how it is taught. Some singing teachers’ confusion of nasality and nasal resonance was discussed (Titze, 1991, 2001; Austin, 1997; Campelo, 2017). Research about the effects of singing tasks on velopharyngeal closure was also examined. Velopharyngeal closure appears to be tighter in singing compared to speech, particularly during pitch ascent (Yanagisawa et al., 1991; Austin, 1997) and singers have a refined ability to tune their velopharyngeal port (Birch et al., 2002; Sundberg et al., 2007). While classical singing training may teach singers to reduce nasality when they sing (Jennings & Kuehn, 2008), other research has shown that in singers, small velopharyngeal openings may contribute to boosting the singer’s formant (Sundberg et al., 2007; Gill et al., 2020). Finally, compared to speech, articulatory movements are often slower in singing, which could reduce some of the effects of nasal co-articulation, as described by Bell- Berti & Krakow (1991). Based on the research discussed in chapters 2 and 3, the first research

102

project explored the effect of altered auditory nasal feedback on the regulation of oral-nasal balance in song. The research expanded on de Boer and Bressmann (2017) and de Boer et al.’s (2019) research investigating the effect of altered auditory nasal feedback on the regulation of oral-nasal balance in speech. 8.2 Project 1: Influence of altered auditory feedback on oral- nasal balance in song 8.2.1 Summary Chapter 4 summarized the first research project in this thesis. The project investigated the effect of increased or decreased nasal signal level auditory feedback on the oral-nasal balance of professional singers in comparison to non-singers during a singing task. Using an experimental repeated measures design, thirty participants (10 female singers, 10 female non-singers, 10 male non-singers) between the ages of 18-35 with a mean age of 24.7 (SD 3.23) sang 38 repetitions of a continuously looped melody that was fed into a mixing console connected to the participants’ headphones. Gradual changes to the nasal channel provided more or less nasal sounding feedback to the participants, whose nasalance scores were measured with the Nasometer 6450. A significant main effect of condition was found, and post-hoc Bonferroni multiple comparison tests revealed that for all of the participants, the nasalance scores for the final baseline and the maximum and minimum nasal feedback conditions were all significantly lower than the first baseline (p < 0.05). No main effects for group or gender were found. 8.2.2 Significance The study demonstrated that during the singing task, both singers and non-singers showed lower nasalance scores in response to increased nasal signal level feedback. This finding confirmed the results of previous speech research (de Boer & Bressmann, 2017; de Boer et al., 2019). When the same singers and non-singers heard their nasal feedback decrease, they also demonstrated lower nasalance scores. This finding was different from the previous observations of de Boer & Bressmann (de Boer & Bressmann, 2017; de Boer et al., 2019). Two possibilities may describe why this happened. Since the results of the experiment also revealed that the baseline 2 and baseline 3 scores were lower than the first baseline score, it is possible that the participants had not completely recovered from the first perturbation and were showing after-effects by the end of the experiment. It is also possible that the perturbed auditory feedback (during the maximum level feedback condition) caused the speakers to temporarily update their internal motor model for speech production and that this feedforward motor planning update lasted for the rest of the

103

experiment (Shiller et al., 2009; Patri et al., 2018). No differences were found between the responses of the professional singers compared to the untrained singers indicating that perhaps the internal models that guide the control of oral-nasal balance may not be as refined in singers as those that guide frequency and intensity (Jones & Keough, 2008; Scheerer & Jones, 2012; Bottalico, Graetzer & Hunter, 2016). 8.2.3 Limitations Previous research (de Boer & Bressmann; 2017 & de Boer et al., 2019) had found no statistically significant result related to feedback condition order (i.e., whether maximum or minimum nasal signal level feedback were presented first), so the present experiment did not permutate the order of the feedback presentation. The increased nasal signal level feedback condition was always presented first, and the decreased nasal signal level feedback condition was always presented second. Moreover, the changes from the maximum and minimum nasal signal levels feedback to baseline were abrupt, while in de Boer and Bressmann (2017) and de Boer et al.’s (2019) research, the auditory perturbations were gradually ramped up and down. Therefore, directly comparing the present results to these studies is challenging, and it is difficult to say whether the changes in experimental design contributed to the following response to the minimum nasal signal level feedback.

In a study by Jennings and Kuehn (2008), classical singers were found to generate lower nasalance scores than amateurs. They suggested that classical singing training may contribute to firmer velopharyngeal closure. No such group difference between the trained and untrained singers was found in the baseline of the first experiment. However, Jennings and Kuehn’s (2008) experiment used sustained vowels, as opposed to the phonetically varied sentence used in this study, which limits direct comparison with their work. Jennings and Kuehn (2008)’s research also recruited solely classical singers, while in the present study, the singers identified with a range of performance styles (i.e., classical, contemporary commercial music, musical theatre, and choral). 8.2.4 Future research perspectives Future research should include a reverse ordering of the perturbation conditions because adaption to feedback can be best studied with abrupt changes from a maximum feedback condition back to baseline (Houde & Jordan, 1998; Jones & Munhall, 2003). In higher registers, velopharyngeal closure has been shown to be firmer in classical singing compared to speech (Yanagisawa et al.,

104

1991; Austin, 1997), so future experiments should also investigate possible differences of the compensation effects in different parts of the singers’ vocal ranges. Finally, it would be interesting to repeat this research with speakers with hypernasality to see whether altered auditory feedback of nasal signal levels could help them reduce their nasalance scores and improve their oral-nasal balance. 8.3 Oral-nasal balance modification Chapter 5 introduced the concept of voice focus (forward and backward) (Boone, 1997; Boone et al., 2010), with a focus on how experts in the field of singing voice pedagogy have understood voice focus related to the bel canto principle of chiaroscuro. Singing voice pedagogy methods to teach focus were discussed, together with research investigating the spectrographic effects of voice focus maneuvers (Sundberg 1974; 1990; Sundberg, Gramming & Lovetri, 1993; Story, Titze & Hoffman, 2001; Stone et al. 2003; Björkner, 2008; Sundberg & Thalén, 2010). How voice focus maneuvers may affect airflow impedance in the vocal tract was also described (Poiseuille, 1846; Sutera & Skalak, 1993; Cole, Forsyth & Haight, 1982; Nair, 2007). Other factors considered were transpalatal sound transfer (Warren, Duany & Fischer, 1969; Mayo, Warren & Zajac, 1998; Gildersleeve-Neumann & Dalston, 2001; Blanton, Watterson & Lewis, 2015), oral sound redirection (Peterson-Falzone, Hardin-Jones & Karnell, 2001; Boone, McFarlane, Von Berg & Zraick, 2010; Kummer, 2014), and vowel effects on nasalance scores (House & Stevens, 1956; Lewis, Watterson & Quint, 2000; Lewis & Watterson, 2003; Awan, Omlor & Watts, 2011; Rong & Kuehn, 2012; Blanton, Watterson & Lewis, 2015; Ha & Cho, 2015). The effect of voice focus adjustments on the regulation of oral-nasal balance in speech and on tongue movement had been investigated in a series of recent studies (Bressmann et al., 2012; de Boer & Bressmann, 2016b; de Boer et al., 2016; Bressmann et al., 2017). Based on these studies, two subsequent experiments examining the effect of forward and backward voice focus adjustments on oral-nasal balance in typical and hypernasal speakers were developed. The studies expanded on previous research (de Boer & Bressmann, 2016b; de Boer et al., 2016) by using singing-voice pedagogy instructional strategies for teaching extreme voice focus adjustments.

105

8.4 Project 2: Influence of voice focus adjustments on oral-nasal balance in speech and song 8.4.1 Summary Chapter 6 described Study 2. The objective of this study was to investigate the effect of extreme forward and backward voice focus adjustments on oral-nasal balance in speech and singing in typical speakers. Using an experimental repeated measures design, twenty participants (10 males, 10 females) between the ages of 18-35 with a mean age of 24.25 (SD 3.73) read balanced, oral and nasal speech stimuli, and sang a song with oral and nasal sounds in forward and backward voice focus conditions. Nasalance scores were measured with a Nasometer 6450 at baseline, during the first voice focus condition, during the second voice focus condition, and upon return to baseline. Half of the participants learned the forward voice focus first and the backward voice focus second, while the other half learned the maneuvers in the opposite order. Results supported previous research (de Boer & Bressmann, 2016b; de Boer et al., 2016) demonstrating that forward voice focus increased nasalance scores for the oral stimulus and song (p < 0.01), and backward voice focus decreased nasalance scores for the nasal stimulus, the phonetically balanced paragraph and the song (p < 0.01). Moreover, it was found that during production of the song, males had higher nasalance scores in the forward voice focus condition than females (45.90% versus 37.00%, p = 0.01). Intriguingly, one female outlier also produced significantly lower nasalance scores in response to the forward voice focus adjustment. Previous research reported that during pitch raising the larynx tends to rise in non-singers (Shipp & Morissey, 1977; Iwarsson & Sundberg, 1998). In trained singers, there have been opposing findings. The larynx has been shown to rise (Shipp & Izdebski, 1975) and fall (Yanagisawa et al., 1998; Echternach et al., 2016). Research has also shown that when loudness is increased, trained singers lower the larynx (Shipp 1984; Echternach et al., 2016). Therefore, the effect of voice focus adjustments on frequency and intensity was also explored. Relative to fundamental frequency, there was a main effect for condition (p < 0.001) and overall, post-hoc tests indicated that the average frequency measures for all of the stimuli were found to be higher in both the forward and backward focus conditions, compared to the baseline conditions (p < 0.05). Across stimuli, females also produced consistently higher frequencies than males, which was to be expected. For intensity, there were significant main effects for condition for the oral (p = 0.01) and nasal stimuli (p = 0.01), and for the song (p = 0.03). Post-hoc tests indicated that for the oral and nasal stimuli, the intensity measurements in the backward voice focus condition were

106

significantly higher than baseline 1 (p < 0.05). Post-hoc tests for the song indicated that the intensity measurements in the forward voice focus condition were significantly higher than baseline 1 (p < 0.05). Pearson correlation coefficients were all very weak (r < 0.2) and mostly statistically non-significant between nasalance, frequency and intensity. 8.4.2 Significance The results of Study 2 supported previous research (de Boer & Bressmann, 2016b; de Boer et al., 2016), demonstrating that voice focus adjustments can influence the oral-nasal balance of typical speakers. Forward voice focus with a raised larynx and shortened vocal tract resulted in more nasalance for the oral stimulus and phonetically balanced song, and backward voice focus with a lowered larynx and lengthened vocal tract resulted in a decrease in nasalance for the nasal stimulus, phonetically balanced paragraph and song. The study also found an outlier whose nasalance scores dropped in response to the forward voice focus condition. In a case study by Bressmann et al., (2012), a speaker with hypernasality also presented with reduced nasalance scores when producing speech with forward voice focus. It could be speculated that the forward voice focus condition may have resulted in a narrower nasopharyngeal constriction for the participant, resulting in improved velopharyngeal closure. During production of the song, males responded to the forward voice focus condition with significantly higher nasalance scores than females. Compared to females, the resting position of the male larynx is lower (Hirano, Kurita & Nakashima, 1983) and the vocal tract length is longer (Fitch & Giedd, 1999). Therefore, it is possible that the relatively lower male larynx had to travel further from its baseline rest position, which may have caused a more substantial reaction to the forward focus adjustment for the men, compared to the women. No other statistically significant differences were found between genders. The effects of voice focus on the song followed a similar pattern to that of the phonetically balanced paragraph. Unfortunately, the phonetic content between the stimuli was too different to compare them statistically. The average measures of fundamental frequency were found to be higher in both the forward and backward focus conditions. Females consistently produced higher frequencies than males, which was expected. The differences in intensity were fairly small overall. The effects of the voice focus adjustments on pitch and intensity were within normative variability ranges for connected speech (Gramming, Sundberg & Akerlund, 1991; Boone et al., 2010), so the overall effect of the adjustments on these parameters appeared to have been limited. Since the correlations between pitch, intensity and nasalance were very

107

weak, the effects on nasalance can be attributed to the voice focus adjustments rather than changes in pitch or intensity. 8.4.3 Limitations The voice focus procedures were taught by a single researcher who also determined when the subjects had achieved the correct vocal tract settings. The phonetic content of the speech and song stimuli were also not matched, limiting the ability to compare them statistically. This was unfortunate given that, compared to the phonetically balanced paragraph, the singing task produced higher numerical scores in the forward voice focus condition (41.45% versus 34.45%). 8.4.4 Future research perspectives The results of Study 2 confirmed that extreme voice focus affects nasalance scores, with relatively minor differences between genders. Future research should incorporate direct visualization of the velopharyngeal sphincter during the voice focus procedures. Most importantly, the voice focus adjustments procedures should be tested with speakers with hypernasality. This was accomplished in the third study. 8.5 Project 3: Immediate effects of voice focus adjustments on hypernasal speakers’ nasalance scores 8.5.1 Summary The objective of the third experiment was to explore how voice adjustments (forward and backward) affected the oral-nasal balance of speakers with hypernasality, measured with nasalance scores. The study was conducted at the Palato-Facial Management Clinic at Five Counties Children’s Centre in Peterborough, Ontario. Using an experimental repeated measures design, five hypernasal children (2 males, 3 females) between the ages of 5-12 with a mean age of 7.4 (SD 2.7) followed the study protocol set out in Study 2, repeating balanced, oral and nasal speech stimuli, and singing a song in both backward and forward voice focus conditions. Descriptive results indicated that for the oral stimulus, forward focus caused the participants’ nasalance scores to increase by 12%. Across all of the speech stimuli, repeated-measures ANOVA findings showed that forward voice focus resulted in a significant increase in mean nasalance compared to baseline 1, backward voice focus and baseline 2 (p < 0.05). Individually, forward voice focus also resulted in considerable increases in nasalance (over 10%) for three participants (i.e., participants 1, 3 and 5). However, in the forward voice focus condition, participant 4’ s nasalance score decreased by -22% during production of the nasal sentence. In backward voice focus, participants’ nasalance scores decreased by 13%. Across all speech

108

stimuli, inferential testing with a repeated measures ANOVA showed that backward voice focus resulted in a significant decrease in mean nasalance compared to baseline 1 and forward voice focus (p < 0.05). Individually, backward voice focus also resulted in significant decreases (equal to or over 10%) for three participants (i.e., participants 1, 4 and 7). 8.5.2 Significance The results of this final project supported previous research for typical speakers (de Boer & Bressmann, 2016b; de Boer et al., 2016) and dovetailed with the findings from study 2 in this thesis (Santoni et al., 2019). Extreme forward voice focus with a raised larynx and shortened vocal tract increased nasalance scores and extreme backward voice focus with a lowered larynx and lengthened vocal tract reduced nasalance scores. Intriguingly, one male participant produced a nasalance score that was -22% lower than his baseline 1 score during the forward voice focus condition. This result lined up with two previously reported cases, one for a female speaker with hypernasality (Bressmann et al., 2012) and one for a female typical speaker (Santoni et al., 2019). It is possible that the vocal tract setting for forward voice focus induced a nasopharyngeal constriction that many have been responsible for the decrease of the nasalance score. Due to the fact that velopharyngeal sphincter morphology can cause speakers to use different closure patterns (Skolnick et al., 1975; Jordan et al., 2017), it is also possible that the effects of voice focus adjustments are individually varied across speakers. This may be even more so the case in speakers with hypernasality who may have additional morphological variation of the velopharyngeal sphincter (Kummer, 2014).

The impedance of airflow in the oral and nasal cavities may have influenced the nasalance scores (Warren, Duany & Fischer, 1969; Mayo, Warren & Zajac, 1998). Previously, it had been proposed that hypernasality could be reduced by rerouting sound through the oral cavity using a mouth-opening maneuver (McDonald & Koepp-Baker, 1951; Shelton et al., 1968; Coston, 1986; Van Demark & Hardin, 1990; Boone & McFarlane, 1994; Peterson-Falzone, Hardin-Jones & Karnell, 2001; Boone, McFarlane, Von Berg & Zraick, 2010; Kummer, 2014). However, while Alighieri and colleagues (2020) reported that an increase in mouth opening reduced nasalance scores in normal speakers, the same was not true for speakers with cleft-palate related hypernasality. Future research should ideally include imaging to document the effect of each voice focus adjustment on the velopharyngeal closing mechanism.

109

8.5.3 Limitations Originally, the study included seven participants. However, analysis of baseline 1 nasalance scores indicated that overall, only five out of the seven participants were clearly in the hypernasal range, presenting with scores equal to or over 28% for the oral sentence (Dalston, Neiman & Gonzalez-Landa, 1993). While the two non-hypernasal participants were moveable from baseline, they had to be removed from the study sample. The remaining sample included only five participants, which was relatively small. In future research, it would be desirable to increase the sample size. Given that in singing, vowels are elongated, which probably reduces their co-articulatory effect on the opening and closing of the velopharyngeal sphincter (Bell-Berti & Krakow, 1991), and given that during pitch ascent, closure of the velopharyngeal sphincter has been reported as firmer in classical singing compared to speech (Yanagisawa et al., 1991; Austin, 1997), a chief objective of this thesis was to understand if voice focus adjustments affect speech and song differently. “The Hamper Song” was designed as an oral-nasal stimulus for experiment 1 (Santoni et al., 2018). For the remaining experiments, the intention was to compare the standard oral-nasal Rainbow Passage (Fairbanks, 1960) to the oral-nasal “Hamper Song”. In hindsight, this was not ideal because the phonetic content of the two stimuli was not the same. In future research, experiments should include a song using text from the Rainbow Passage (Fairbanks, 1960) or a spoken version of “The Hamper Song” text.

The assessment of the speakers’ effectiveness of attaining forward and backward focus was based solely on nasalance scores. However, correlations between nasalance scores and auditory- perceptual nasality assessment have been documented as variable across the literature (Dalston, Warren & Dalston, 1991; Hardin et al., 1992; Dalston, Neiman & Gonzalez-Landa, 1993; Keuning et al., 2002; Brunnegård, Lohmander & van Doorn, 2012). Since listener perception of hypernasality can be variable and listener consensus can be poor (Keuning et al., 2002), assessing nasality perceptually during each of the voice focus adjustments would likely have been very challenging. Nevertheless, auditory-perceptual analyses should be included in future work. Finally, even though this study has shown that extreme voice focus adjustments have potential as facilitating techniques to improve nasalance scores, the extreme postures may not result in ideal long-term voice qualities. Strategies to gradually phase out the postures while maintaining improved nasalance scores need to be explored.

110

8.5.4 Future research perspectives The results of Study 3 demonstrated that, for the whole group of five speakers with hypernasality, forward voice focus increased, and backward voice focus decreased nasalance scores. There was one exception. Forward voice focus led to a decrease in nasalance for a male participant. Future research should examine the effectiveness of the extreme voice focus intervention as a potential therapy technique for select speakers with hypernasality. Such future research should ideally also include velopharyngeal imaging and detailed auditory-perceptual assessments. It would also be important to explore long-term effects and retention of the intervention. 8.6 Conclusion and future directions

This dissertation investigated how to influence oral-nasal balance in speech and song. Literature was reviewed from the fields of speech science and singing voice pedagogy, which inspired three dissertation projects. Collectively, the projects answered several questions. Study 1 established the role of perturbed auditory nasal signal level feedback on the control of oral-nasal balance during a singing task. Studies 2 and 3 confirmed that for typical (study 2) and hypernasal (study 3) speakers, oral-nasal balance is influenced by how the vocal tract is aligned (i.e., voice focus: a training technique from singing voice pedagogy). Directions for future study should include determining how nasal feedback manipulation affects the oral-nasal balance of speakers with hypernasality in speech and song. Future clinical research on the voice focus intervention is also needed in order to draw more definitive conclusions on its therapeutic potential for speakers with hypernasality. While the research provided in this dissertation may appear as somewhat divergent, the interventions can be merged in future research. It would be intriguing to investigate a sequenced or combined intervention that involves manipulating an individual’s auditory-perceptual environment while adjusting their vocal tract settings and to observe the combined effect on oral-nasal balance

Overall, the findings in this dissertation suggest a number of additional avenues for future research and clinical integration. In order to further refine the voice focus intervention, it may be beneficial to analyze its laryngeal component and articulatory component independently. From this vantage point, two paths for further research emerge from the literature.

111

The first idea is specific to the laryngeal aspect of voice focus adjustment. Classical singers have been described as having a refined ability towards pre-phonatory tuning of the larynx (Shipp et al., 1987; Pabst & Sundberg, 1992; Aura et al., 2019). In order to maintain a comfortably low laryngeal position, the concept of singing “on the gesture of a yawn” has been promoted across the literature (Garcia, 1894; McCoy, 2004; McKinney, 2005; Bozeman, 2013). CCM singers have also been shown to pre-set the larynx (Estill, 1988; Bourne & Garnier, 2012). It may be possible that with practice (as has been shown in singing), adjusting a speakers’ global laryngeal position (i.e., the laryngeal component of the voice focus adjustment) could be learned over time. A more longitudinal voice focus study is recommended to explore this possibility.

The second idea is specific to the articulatory component of the voice focus adjustment. Formant tuning is another example of manipulating the vocal tract, which Sundberg (1988) describes as highly dependent on tongue shaping. In a recent formant analyses study of indie-pop singers by Jones, Schellenberg and Gick (2017), a significant difference in F1 was found between sung and spoken versions of the vowel /i/ (p = 0.03) wherein F1 was higher in singing compared to speech. The authors proposed that this was due to the pharyngeal constriction associated with the indie-pop singing style. They further speculated that in the indie-pop style, /i/ is likely displaced due to the stylistic front-rising diphthongs characteristic of the genre (with front rising diphthongs, the tongue glides from a low position to a high position [e.g., /aɪ/, /eɪ/, or /ɔɪ/]) (Burridge & Stebbins, 2016). This dependency on adjacent speech sounds can also be connected to research findings by McIver and Miller (1995). Their work demonstrated that in singing, vowels preceding the nasal /m/ produced higher nasalance scores than vowels following /m/. Their results implied that the effect of anticipatory nasality may be greater than carryover nasality in singing. In future voice focus research, the same principles may serve to improve maintenance of the forward and backward vocal tract settings. Since the forward voice focus protocol employs use of the vowel /i/, and the backward voice focus protocol employs use of the vowel /a/ (i.e., the articulatory component of the voice focus adjustment), vocal tract setting preservation might be increased if in future work, participants were asked to include gestural diphthongs bookending each sound. Since previous research literature has reported that singing style is largely informed by vocal tract settings (Bateman, 2003; Nair, Nair & Reishofer, 2016; Jones et al.’s (2017), singing in a specific style could hypothetically induce the same effect (i.e., indie-pop vs. operatic). By breaking down the voice focus maneuvers into two smaller

112

components (e.g., laryngeal and articulatory) and practising them independently using the strategies as described above, it may be possible to maximize the intervention effects.

Finally, in an effort to conceptualize how the results of the first experiment might be combined with the results from the second and third experiments, a study on the effect of room reverberation from the singing voice pedagogy literature offers some first clues. Ternström (1993) conducted an experiment on three different choral groups (a boys choir, a mixed youth choir and a mixed adult choir) singing in three different acoustic spaces (basement, hall and church), graded by their reverberation times (measured as the time it takes for sound to decay by 60dB). The author investigated the effect of room acoustics, vocal effort, musical material and type of choir on long term average spectrum measurements. Irrespective of musical material or choir identity, the results indicated 5-10% higher formant frequencies for all three choirs in the basement room, which had shorter reverb. The author speculated that the adaptation of vocal effort to room acoustics via an elevated laryngeal position (shortened vocal tract) may have been responsible for this outcome because shortening the vocal tract position produces higher formant frequencies. It is well known that singers adjust their performance to the acoustic environment. This affects tempo, vibrato, intensity (Ueno, Kato & Kawai, 2010), musical articulation (Kato, Ueno & Kawai, 2015) and timbre (Luizard et al., 2018; Luizard, Steffens & Weinzierl, 2020). Luizard et al. (2018) demonstrated that timbre adaption to acoustic conditions could be measured using spectral flux, i.e., how quickly the power spectrum changes. All of this research offers clues regarding how an individual’s acoustical environment could contribute to maintenance of the voice focus postures and affect oral-nasal balance.

The findings in this dissertation may also provide an opportunity to interface with other music- inspired rehabilitation approaches, such as Neurologic Music Therapy (NMT, Thaut & Hoemberg, 2014). NMT is defined as “…the therapeutic application of music to cognitive, sensory and motor dysfunctions due to neurologic disease of the human nervous system.” (Thaut, 2014, pp. 2). NMT has twenty therapeutic music exercises, which do not exist as a complimentary enhancement to therapy, but as a means for functional rehabilitation in the areas of cognition, sensorimotor control, as well as speech and language (Thaut, 2005). One example speech and language exercise is Vocal Intonation Therapy (VIT). VIT employs the use of vocal exercises to “…train, maintain, develop, and rehabilitate aspects of voice control due to

113

structural, neurological, physiological, psychological, or functional abnormalities of the voice apparatus.” (Thaut, 2014, pp. 179). The voice focus intervention could potentially be adapted into a VIT technique for treating individuals with symptomatic hypernasality of different structural or neurological etiologies.

114

References

Alderson, A. (1993). Positioning the velum. Journal of Research in Singing and Applied Vocal Pedagogy, 16(2), 25-32. Alighieri, C., Bettens, K., Everaert, J., & Van Lierde, K. (2020). Impact of speech rate and mouth opening on hypernasality and speech intelligibility in children with a cleft (lip and) palate. Journal of Communication Disorders, 83, 105950. doi: 10.1016/j.jcomdis.2019.105950. Allori, A.C., Kelley, T., Meara, J.G., Albert, A., Bonanthaya, K., Chapman, K., Cunningham, M., Daskalogiannakis, J., de Gier, H., Heggie, A.A., Hernandez, C., Jackson, O., Jones, Y., Kangesu, L., Koudstaal, M.J., Kuchhal, R., Lohmander, A., Long, R.E. Jr., Magee, L., Monson, L., Rose, E., Sitzman, T.J., Taylor, J.A., Thorburn, G., van Eeden, S., Williams, C., Wirthlin, J.O., & Wong, K. (2017). A standard set of outcome measures for the comprehensive appraisal of cleft care. The Cleft Palate - Craniofacial Journal, 54(5), 540-544. doi: 10.1597/15-292 American Academy of Music Teachers of Singing. (2008). In support of Contemporary Commercial Music (nonclassical) voice pedagogy. Journal of Singing, 65(1), 7-10. Aura, M., Geneid, A., Bjørkøy, K., Rantanen, M., & Laukkanen, A. M. (2019). The nasal musculature as a control panel for singing - why classical singers use a special facial expression. Journal of Voice, 33(4), 510-515. doi:10.1016/j.jvoice.2017.12.016 Austin, S. F. (2000). From song to speech: Nasal resonance - fact or fiction? Journal of Singing - the Official Journal of the National Association of Teachers of Singing, 57(2), 33-40. Austin, S.F. (1997). Movement of the velum during speech and singing in classically trained singers. Journal of Voice, 11(2), 212-221. doi:10.1016/S0892-1997(97)80080-X Awan, S. N., Omlor, K., & Watts, C. R. (2011). Effects of computer system and vowel loading on measures of nasalance. Journal of Speech, Language, and Hearing Research, 54(5), 1284- 1294. doi: 10.1044/1092-4388(2011/10-0201) Bailey, R. (1993). Electronic methods for treating nasality in classical singing. Journal of Research in Singing and Applied Vocal Pedagogy, 17(1), 31-48. Balog, J. E. (2005). Popular song and music theater: A guide to evaluating music theater singing for the classical teacher. Journal of Singing - the Official Journal of the National Association of Teachers of Singing, 61(4), 401-406. Bateman, L.A. (2003). Soprano, style and voice quality: acoustic and laryngographic correlates. (Unpublished master’s thesis). University of Victoria, Victoria, BC.

115

Bauer, J. J., Mittal, J., Larson, C. R., & Hain, T. C. (2006). Vocal responses to unanticipated perturbations in voice loudness feedback: An automatic mechanism for stabilizing voice amplitude. The Journal of the Acoustical Society of America, 119(4), 2363-2371. doi: 10.1121/1.2173513 Baylis, A., Chapman, K., & Whitehill, T. L. (2015). Validity and reliability of visual analog scaling for assessment of hypernasality and audible nasal emission in children with repaired cleft palate. The Cleft Palate - Craniofacial Journal, 52(6), 660-670. doi: 10.1597/14-040 Baylis, A.L., Munson, B., & Moller, K.T. (2011). Perceptions of audible nasal emission in speakers with cleft palate: A comparative study of listener judgments. The Cleft Palate - Craniofacial Journal, 48(4), 399-411. doi: 10.1597/09-201 Beaty, T. H., Ruczinski, I., Murray, J. C., Marazita, M. L., Munger, R. G., Hetmanski, J. B., Murray, T., Redett, R. J., Fallin, M. D., Liang, K. Y., Wu, T., Patel, P. J., Jin, S., Zhang, T. X., Schwender, H., Wu‐Chou, Y. H., Chen, P. K., Chong, S. S., Cheah, F., … Scott, A. F. (2011). Evidence for gene‐environment interaction in a genome wide study of nonsyndromic cleft palate. Genetic Epidemiology, 35(6), 469–478. doi: 10.1002/gepi.20595 Behrman, A. (2013). Speech and voice science (2nd ed.). San Diego, CA: Plural Publishing. Behrman, A. and Haskell, J. (Eds.). (2013) Exercises for voice therapy (2nd ed.). San Diego, CA: Plural Publishing Inc. Bell-Berti, F., & Krakow, R. (1991). Anticipatory velar lowering: a coproduction account. The Journal of the Acoustical Society of America, 90(1), 112-123. doi:10.1121/1.401304 Bestebreurtje, M. E., & Schutte, H. K. (2000). Resonance strategies for the belting style: Results of a single female subject study. Journal of Voice, 14(2), 194-204. doi:10.1016/S0892- 1997(00)80027-2 Bettens, K., Van Lierde, K.M, Corthals, P., Luyten, A., & Wuyts, F. L. (2016). The nasality severity index 2.0: Revision of an objective multiparametric approach to hypernasality. The Cleft Palate - Craniofacial Journal, 53(3), E60-E70. doi: 10.1597/14-247 Bettens, K., Wuyts, F. L., & Van Lierde, K. M. (2014). Instrumental assessment of velopharyngeal function and resonance: a review. Journal of Communication Disorders, 52, 170-183. doi:10.1016/j.jcomdis.2014.05.004 Billmire, D. A. (2008). Surgical management of clefts and velopharyngeal dysfunction. In A. Kummer (Ed.), Cleft palate and craniofacial anomalies – Effects on speech and resonance (2nd ed., pp. 508-540). New York, NY: Delmar Cengage Learning.

116

Birch, P., Gümoes, B., Stavad, H., Prytz, S., Björkner, E., & Sundberg, J. (2002). Velum behavior in professional classic operatic singing. Journal of Voice, 16(1), 61-71. doi:10.1016/S0892- 1997(02)00073-5 Björkner, E. (2008). Musical theater and opera singing - why so different? A study of subglottal pressure, voice source, and formant frequency characteristics. Journal of Voice - Official Journal of the Voice Foundation, 22(5), 533-540. doi: 10.1016/j.jvoice.2006.12.007 Blanton, A., Watterson, T., & Lewis, K. (2015). The differential influence of vowels and palatal covering on nasalance scores. The Cleft Palate-Craniofacial Journal, 52(1), 82-87. doi: 10.1597/13-092 Bloch, G. W. (2007). The pathological voice of Gilbert-Louis Duprez. Cambridge Opera Journal, 19(1), 11-31. doi:10.1017/S0954586707002248 Bloem-Hubatka, D. (2012). The old Italian school of singing: A theoretical and practical guide. Jefferson, NC & London, UK: McFarland & Company Inc. Blood, G. W., & Hyman, M. (1977). Children’s perception of nasal resonance. Journal of Speech and Hearing Disorders, 42(3), 446-448. doi: 10.1044/jshd.4203.446 Boone D.R. (1997). Is your voice telling on you? (2nd ed.). San Diego, CA: Singular. Boone, D. R., & McFarlane, S. C. (1993). A critical view of the yawn-sigh as a voice therapy technique. Journal of Voice, 7(1), 75–80. doi: 10.1016/S0892-1997(05)80114-6 Boone, D. R., McFarlane, S. C., Von Berg, S. L., & Zraick, R. I. (2010). The voice and voice therapy (8th ed.). New York, NY: Allyn & Bacon. Boone, D.R., & McFarlane, S.C. (1988). The voice and voice therapy (4th ed.). Englewood Cliffs, NJ: Prentice-Hall. Boone, D.R., & McFarlane, S.C. (1994). The voice and voice therapy (5th ed.). Englewood Cliffs, NJ: Prentice-Hall. Borg, E. (2007). On Perceived Exertion and its Measurement. (Unpublished doctoral dissertation), Stockholm, Sweden: Stockholm University. Borg, G. (1982). A category scale with ratio properties for intermodal and interindividual comparisons. In H.G. Geissler, & P. Petzold (Eds.), Psychophysical Judgment and the Process of Perception (pp. 25-34). Berlin, Germany: Deutscher Verlag der Wissenschaften. Borg, G., & Borg, E. (2001). A new generation of scaling methods: level-anchored ratio scaling. Psychologica 28, 15-45.

117

Bottalico, P., Graetzer, S., & Hunter, E. J. (2016). Effect of training and level of external auditory feedback on the singing voice: Volume and quality. Journal of Voice, 30(4), 434-442. doi:10.1016/j.jvoice.2015.05.010 Bourne, T., & Garnier, M. (2012). Physiological and acoustic characteristics of the female Music Theater voice. Journal of the Acoustical Society of America, 131(2), 1586-1594. doi:10.1121/1.3675010 Bozeman, K. (2013). Practical vocal acoustics: pedagogic applications for teachers and singers. Hillsdale, NY: Pendragon Press. Bressmann, T., Anderson, J. D., Carmichael, R. P., & Mellies, C. (2012). Prosthodontic management of hypernasality: Two very different cases/Gestion de l'hypernasalite a l'aide de protheses velopalatines: deux cas tres differents. Canadian Journal of Speech-Language Pathology and Audiology, 36(1), 50-57. Bressmann, T., de Boer, G., Marino, V. C. de C., Fabron, E. M. G., & Berti, L. C. (2017). Influence of voice focus on tongue movement in speech. Clinical Linguistics & Phonetics, 31(3), 212–221. doi: 10.1080/02699206.2016.1235228 Broomhead, I.W. (1951). The nerve supply of the muscles of the soft palate. Br J Please Surg 4, 1-15. Broomhead, I.W. (1957). The nerve supply of the soft palate. Br J Plast Surg 10, 81-88. Bruneel, L., Bettens, K., De Bodt, M., D’haeseleer, E., Thijs, Z., Roche, N., & Van Lierde, K. (2020). Stages in the development and validation of a Belgian Dutch outcome tool for the perceptual evaluation of speech in patients with cleft palate. The Cleft Palate-Craniofacial Journal, 57(1), 43–54. doi: 10.1177/1055665619862726 Brunnegård, K., Lohmander, A., & Van Doorn, J. (2012). Comparison between perceptual assessments of nasality and nasalance scores. International Journal of Language and Communication Disorders, 47(5), 556-566. doi: 10.1111/j.1460-6984.2012.00165.x. Burnett, T. A., Freedland, M. B., Larson, C. R., & Hain, T. C. (1998). Voice F0 response to manipulations in pitch feedback. Journal of the Acoustical Society of America, 103, 3153-3161. doi:10.1121/1.423073 Burridge, K., & Stebbins, T. (2016). For the love of language: An introduction to linguistics. Cambridge, UK: Cambridge University Press. Bzoch, K.R. (1968). Variations in velopharyngeal valving: the factor of vowel changes. The Cleft Palate Journal, 5, 211-218.

118

Bzoch, K.R. (1997). Introduction to the study of communication disorders in cleft palate and related craniofacial anomalies. In K.R. Bzoch (Ed.), Communicative disorders related to cleft lip and palate (4th ed., pp. 3-44). Austin, Tx: Pro-Ed. Caleon, I., & Ramanathan, S. (2008). From music to physics: The undervalued legacy of Pythagoras. Science & Education, 17(4), 449-456. doi:10.1007/s11191-007-9090-x Callaghan, J. (2000). Singing and voice science. San Diego, CA: Singular Pub. Group. Campelo, A. (2017). Singing Portuguese nasal vowels: Practical strategies for managing nasality in Brazilian art songs (Unpublished doctoral dissertation). University of Kentucky, Lexington, Kentucky. Cassell, M.D., & Elkadi, H. (1995). Anatomy and physiology of the palate and velopharyngeal structures. In R.J. Shprintzen & J. Bardach (Eds.), Cleft palate speech management: a multidisciplinary approach (pp. 45-62). St Louis, MO: Mosby. Chang-Yit, R., Pick Jr, H. L., & Siegel, G. M. (1975). Reliability of sidetone amplification effect in vocal intensity. Journal of Communication Disorders, 8(4), 317-324. doi:10.1016/0021- 9924(75)90032-5 Chapman, K. L., Baylis, A., Trost-Cardamone, J., Cordero, K. N., Dixon, A., Dobbelsteyn, C., Thurmes, A., Wilson, K., Harding-Bell, A., Sweeney, T., Stoddard, G., & Sell, D. (2016). The Americleft Speech Project: A training and reliability study. The Cleft Palate - Craniofacial Journal, 53(1), 93-108. Doi: 10.1597/14-027 Chapman, K. L., Hardin-Jones, M., & Halter, K. A. (2003). The relationship between early speech and later speech and language performance for children with cleft lip and palate. Clinical Linguistics and Phonetics, 17(3), 173-197. doi:10.1080/0269920021000047864 Clarke, W.M., & Mackiewicz-Krassowska, H. (1977). Variation in the oral and nasal sound pressure level of vowels in changing phonetic contexts. Journal of Phonetics, 5(3), 195-203. doi: 10.1016/S0095-4470(19)31134-9. Coffin, B. (1980). Overtones of Bel Canto. Metuchen, NJ: Scarecrow. Coffin, B. (1987). Coffin's sounds of singing: Principles and applications of vocal techniques with chromatic vowel chart. Metuchen, NJ: Scarecrow Press. Coffin, B. (1989). Historical vocal pedagogy classics. Metuchen, NJ: Scarecrow Press. Cohen, N.S. (1994). Speech and Song: Implications for Therapy. Music Therapy Perspectives, 12(1), 8– 14. doi: 10.1093/mtp/12.1.8

119

Cole, P., Forsyth, R., & Haight, J.S.J. (1982). Respiratory resistance of the oral airway. Am Rev Respir Dis, 125(3), 363-365. doi: 10.1164/arrd.1982.125.3.363 Collett, B. R., & Speltz, M. L. (2007). A developmental approach to mental health for children and adolescents with orofacial clefts. Orthod Craniofac Res, 10(3), 138-148. doi:10.1111/j.1601- 6343.2007.00394.x Cooper, M. & Cooper, M.H. (1977). Approaches to Vocal Rehabilitation. Springfield, IL: Charles C. Thomas. Coston, G.N. (1986). Therapeutic management of speech disorders associated with velopharyngeal incompetence. Journal of Childhood Communication Disorders, 10(1), 75–85. doi: 10.1177/152574018601000107. Craik, J. (2011). Mathilde Marchesi: a study of her life and work in vocal pedagogy, including historical and modern implications (Unpublished doctoral dissertation). Florida Atlantic University, Boca Raton, Florida. Croft, C.B., Shprintzen, R.J., & Rakoff, .SJ. (1981). Patterns of velopharyngeal valving in normal and cleft palate subjects: a multi-view videofluoroscopic and nasendoscopic study. Laryngoscope 91(2), 265–271. doi: 10.1288/00005537-198102000-00015 Curtis, H. H. (1901). Voice building and tone placing: Showing a new method of relieving injured vocal cords by tone exercises. London, UK: Kimpton. Curtis, H.H. (1909). Voice building and tone placing (3rd ed.). London, UK: J. Curnew & Sons. Dalgorf, D.M., & Harvey, R.J. (2013). Chapter 1: Sinonasal anatomy and function. Am J Rhinol Allergy, 27(3), S3-S6. doi: 10.2500/ajra.2013.27.3888 Dalston, R. M., Neiman, G. S., & Gonzalez-Landa, G. (1993). Nasometric sensitivity and specificity: A cross-dialect and cross-culture study. The Cleft Palate-Craniofacial Journal, 30(3), 285–291. doi: 10.1597/1545-1569_1993_030_0285_nsasac_2.3.co_2 Dalston, R.M. (1989). Using simultaneous photodetection and nasometry to monitor velopharyngeal behavior during speech. Journal of Speech and Hearing Research, 32(1), 195-202. doi: 10.1044/jshr.3201.195. Dalston, R.M., Warren, D.W., & Dalston, E.T. (1991). Use of nasometry as a diagnostic tool for identifying patients with velopharyngeal impairment. The Cleft Palate-Craniofacial Journal, 28(2), 184-189. doi: 10.1597/1545-1569_1991_028_0184_uonaad_2.3.co_2. Davids, J., & LaTour, S.A. (2012). Vocal technique: A guide for conductors, teachers, and singers. Long Grove, IL: Waveland Press.

120

De Boer, G., & Bressmann, T. (2014). Comparison of nasalance scores obtained with the Nasometers 6200 and 6450. The Cleft Palate-Craniofacial Journal, 51, 90-97. doi: 10.1597/12-202 De Boer, G., & Bressmann, T. (2015). Application of linear discriminant analysis to the nasometric assessment of resonance disorders: A pilot study. The Cleft Palate-Craniofacial Journal, 52, 173-182. doi: 10.1597/13-109 De Boer, G., & Bressmann, T. (2016a). Application of linear discriminant analysis to the longterm averaged spectra of simulated disorders of oral-nasal balance. The Cleft-Palate Craniofacial Journal, 53(5), e163-e171. doi: 10.1597/14-236 De Boer, G., & Bressmann, T. (2016b). Influence of Voice Focus on Oral-Nasal Balance in Speech. Journal of Voice, 30(6), 705-710. doi:10.1016/j.jvoice.2015.08.021 De Boer, G., & Bressmann, T. (2017). Influence of Altered Auditory Feedback on Oral-Nasal Balance in Speech. Journal of Speech Language and Hearing Research, 60(11), 3135-3143. doi:10.1044/2017_JSLHR-S-16-0390 De Boer, G., Marino, V. C., Berti, L. C., Fabron, E. M., & Bressmann, T. (2016). Influence of Voice Focus on Oral-Nasal Balance in Speakers of Brazilian Portuguese. Folia Phoniatr Logop, 68(3), 152-158. doi:10.1159/000452245 De Boer, G., Marino, V. C., Berti, L. C., Fabron, E. M., Spazzapan, E.A., & Bressmann, T. (2019). Influence of altered auditory feedback on oral-nasal balance in speakers of Brazilian Portuguese. Journal of Speech Language and Hearing Research, 62(10), 3752-3762. doi: 10.1044/2019_JSLHR-S-18-0051 Delattre, P., Liberman, A. M., Cooper, F. S., & Gerstman, L. J. (1952). An experimental study of the acoustic determinants of vowel color; observations on one- and two-formant vowels synthesized from spectrographic patterns. Word, 8(3), 195-210. Devani, P., Watts, R., & Markus, A.F. (1999). Speech outcome in children with cleft palate: Aerophonoscope assessment of nasal emission. Journal of Cranio-Maxillofacial Surgery, 27(3), 180–186. doi: 10.1016/S1010-5182(99)80048-1 Dickson, D.R. (1962). An acoustic study of nasality. Journal of Speech and Hearing Research, 5, 103- 111. Dixon, M., Marazita, M., Beaty, T., & Murray, J. (2011). Cleft lip and palate: understanding genetic and environmental influences. Nat Rev Genet 12, 167–178. doi:10.1038/nrg2933 Domenech-Ratto, G. (1977). Development and peripheral innervation of the palate muscles. Acta Anat 97, 4-14.

121

Doscher, B. (1994). The functional unity of the singing voice (2nd ed.). Lanham, MD: The Scarecrow Press. Duffy, J. R. (2012). Motor speech disorders: substrates, differential diagnosis, and management (3rd ed.). St. Louis, Mo: Elsevier Mosby. Dworkin, J.P., Marunick, M.T., & Krouse, J.H. (2004). Velopharyngeal dysfunction speech characteristics, variable etiologies, evaluation techniques, and differential treatments. Lang Speech Hear Serve Sch, 35(4), 333-352. doi: 10.1044/0161-1461(2004/033) Echternach, M., Burk, F., Burdumy, M., Traser, L., & Richter, B. (2016). Morphometric Differences of Vocal Tract Articulators in Different Loudness Conditions in Singing. PLoS One, 11(4), e0153792. doi:10.1371/journal.pone.0153792 Echternach, M., Popeil, L., Traser, L., Wienhausen, S., & Richter, B. (2014). Vocal tract shapes in different singing functions used in musical theater singing-a pilot study. Journal of Voice, 28(5), 653.e1-653.e7. doi:10.1016/j.jvoice.2014.01.011 Ellis, A.J. (1877). Pronunciation for singers. London, UK: J. Curwen & Sons. Elman, J. L. (1981). Effects of frequency-shifted feedback on the pitch of vocal productions. The Journal of the Acoustical Society of America, 70(1), 45–50. doi: 10.1121/1.386580 Estill, J. (1988). Belting and classic voice quality: some physiological differences. Medical Problems of Performing Artists, 3, 37-43. Fagnan, L. (2008). Chiaroscuro resonance balancing: The bel canto answer to choral tone and intonation problems. The Choral Journal, 49(5), 51-55, 58-61. Fairbanks, G. (1960). Voice and articulation drillbook (2nd ed.). New York, NY: Harper & Row. Fant, G. (1960). Acoustic theory of speech production: With calculations based on X-ray studies of Russian articulations. s'Gravenhage, NL: Mouton. Finkelstein, Y., Lerner, M. A., Ophir, D., Nachmani, A., Hauben, D. J., & Zohar, Y. (1993). Nasopharyngeal profile and velopharyngeal valve mechanism. Plastic and Reconstructive Surgery, 92(4), 603-614. Finkelstein, Y., Shapiro-Feinberg, M., Talmi, Y. P., Nachmani, A., Derowe, A., & Ophir, D. (1995). Axial Configuration of the Velopharyngeal Valve and Its Valving Mechanism. The Cleft Palate- Craniofacial Journal, 32(4), 299–305. doi: 10.1597/1545- 1569_1995_032_0299_acotvv_2.3.co_2

122

Fitch, W. T., & Giedd, J. (1999). Morphology and development of the human vocal tract: a study using magnetic resonance imaging. Journal of the Acoustical Society of Amercia, 106(3 Pt 1), 1511- 1522. doi: 10.1121/1.427148 Flanagan, J. L. (1972). Speech analysis; synthesis and perception. Berlin, DE: Springer-Verlag. Fletcher, S. G. (1970). Theory and instrumentation for quantitative measurement of nasality. The Cleft Palate Journal, 7, 601–609 Fletcher, S. G., & Bishop, M. E. (1970). Measurement of nasality with Tonar. Cleft Palate Journal, 7, 610–621. Fletcher, S. G., Adams, L., & McCutcheon. (1989). Cleft palate speech assessment through oral-nasal acoustic measures. In K. R. Bzoch (Ed.), Communicative disorders related to cleft lip and palate (pp. 246–257). Boston, MA: College Hill Press. Fletcher, S.G. (1976). Nasalance vs. listener judgements of nasality. The Cleft Palate Journal, 13(1), 31-44. Flowers, C.R., & Morris, H.L. (1973). Oral-pharyngeal movements during swallowing and speech. The Cleft Palate Journal, 10(2), 181-191. Fowler, L. P., & Morris, R.J. (2007). Comparison of fundamental frequency nasalance between trained singers and nonsingers for sung vowels. Annals of Otology, Rhinology & Laryngology, 116(10), 739-746. doi: 10.1177/000348940711601005 Ganofsky, M. (2014), Libertine Clairs‐Obscurs: The Enticement of the Shadows. Journal of 18th- Century Studies, 37, 499-515. doi:10.1111/1754-0208.12203 Garcia, M. (1840). Mémoire sur la voix humaine présenté à l’académie des sciences en 1840. Paris, FR: Surverger. Garcia, M. (1894). Hints on singing. London, UK: Ascherberg, Hopwood and Crew. Gart, M. S., & Gosain, A. K. (2014). Surgical management of velopharyngeal insufficiency. Clinical Plastic Surgery, 41, 253-270. doi: 10.1016/j.cps.2013.12.010 Gildersleeve-Neumann, C. E., & Dalston, R. M. (2001). Nasalance scores in noncleft individuals: Why not zero? The Cleft Palate-Craniofacial Journal, 38, 106-111. doi: 10.1597/1545- 1569(2001)038<0106:NSINIW>2.0.CO;2 Gill, G.P., Lee, J., Lã, F.M.B., & Sundberg, J. (2020). Spectrum effects of a velopharyngeal opening in singing. Journal of Voice, 34(3), 346-351. doi: 10.1016/j.jvoice.2018.11.014

123

Glaze, L.E. (2009). Behavioral approaches to treating velopharyngeal dysfunction and nasality. In K.T. Moller & L.E. Glaze (Eds.), Cleft palate: Interdisciplinary issues and treatment (2nd ed., pp. 415-452). Austin, TX: Pro-Ed. Golding-Kushner, K. J. (2001). Therapy techniques for cleft palate speech and related disorders. Clifton Park, NY: Delmar Learning. Golding-Kushner, K.J. (1995). Treatment of articulation and resonance disorders associated with cleft palate and VPI. In R.J. Shprintzen & J. Bardach (Eds.), Cleft palate speech management: A multidisciplinary approach (pp. 327-351). St. Louis, MO: Mosby. Goldstein, E.B. (2002). Sensation and perception (6th ed.). Australia: Wadsworth-Thomson Learning. Gosain, A.K., Conley, S.F., Marks, S., & Larson, D.L. (1996). Submucous cleft palate: diagnostic methods and outcomes of surgical treatment. Plast Reconstr Surg., 97, 1497-1509. Gramming, P., Nord, L., Sundberg, J., & Elliot, N. (1993). Does the nose resonate during singing?. STL- QPSR, 34(4), 35-42. Gramming, P., Sundberg, J., & Akerlund, L. (1991). Variability of phonetograms. Folia Phoniatrica, 43(2), 79-92. doi: 10.1159/000266116 Granqvist, S. (2003). The visual sort and rate method for perceptual evaluation in listening tests. Logopedics Phonatrics Vocology, 28(3), 109–116. doi: 10.1080/14015430310015255 Guenther, F. H. (1995). Speech sound acquisition, coarticulation, and rate effects in a neural network model of speech production. Psychological Review, 102(3), 594-621. doi: 10.1037/0033- 295X.102.3.594 Guenther, F. H. (2006). Cortical interactions underlying the production of speech sounds. Journal of Communication Disorders, 39(5), 350-365. doi: 10.1016/j.jcomdis.2006.06.013 Guenther, F. H., & Vladusich, T. (2012). A neural theory of speech acquisition and production. Journal of Neurolinguistics, 25(5), 408-422. doi:10.1016/j.jneuroling.2009.08.006 Guenther, F.H. (1995). A modeling framework for speech motor development and kinematic articulator control. Proceedings of the XIIIth International Congress of Phonetic Sciences 2, 92-99. Guzman, M., Lanas, A., Olavarria, C., Azocar, M. J., Muñoz, D., Madrid, S., Monsalve, S., Martinez, F., Vargas, S., Cortez, P., & Mayerhoff, R. M. (2015). Laryngoscopic and spectral analysis of laryngeal and pharyngeal configuration in non-classical singing styles. Journal of Voice, 29(1), 130.e21-130.e28. doi:10.1016/j.jvoice.2014.05.004

124

Ha, S., & Cho, S. (2015). Nasalance scores for normal Korean-speaking adults and children: Effects of age, vowel context, and stimulus length. International Journal of Pediatric Otorhinolaryngology, 79(8), 1235-1239. doi: 10.1016/j.ijporl.2015.05.019. Hanrahan, K. (2012). Effect of auditory feedback on singing. Journal of Singing, 69(2), 145-152. Hardin-Jones, M. A., & Jones, D. L. (2005). Speech production of preschoolers with cleft palate. The Cleft Palate-Craniofacial Journal, 42(1), 7-13. doi: 10.1597/03-134.1 Hardin, M.A., Van Demark, D.R., Morris, H.L., & Payne, M.M. (1992). Correspondence between nasalance scores and listener judgments of hypernasality and hyponasality. The Cleft Palate- Craniofacial Journal, 29(4), 346–351. doi: 10.1597/1545- 1569_1992_029_0346_cbnsal_2.3.co_2. Harding, A., Harland, K. and Razzell, R. (1997). Cleft Audit Protocol for Speech (CAPS). Available from Speech/Language Therapy Department, St Andrews Centre for Plastic Surgery Centre, Broomfield Hospital, Court Road, Broomfield, Chelmsford, Essex, UK. Hartnick, C.J., & Boseley, M.E. (2010). Clinical management of children's voice disorders. San Diego, CA: Plural Publishing. Helmholtz, H.V. (1885). On the sensations of tone as a physiological basis for the theory of music (2nd English translation by Alexander John Ellis). London, UK: Longmans, Green and Co. Hemsley, T. (1998). Singing and imagination: A human approach to a great musical tradition. Oxford, UK: Oxford University Press. Henningsson, G., Kuehn, D.P., Sell, D., Sweeney, T., Trost-Cardamone, J., & Whitehill, T.L. (2008). Universal parameters for reporting speech outcomes in individuals with cleft palate. The Cleft Palate-Craniofacial Journal, 45(1), 1-17. doi: 10.1597/06-086.1 Hickok, G., Houde, J., & Rong, F. (2011). Sensorimotor integration in speech processing: Computational basis and neural organization. Neuron, 69(3), 407-22. doi: 10.1016/j.neuron.2011.01.019 Hirano, M., Kurita, S., Nakashima, T. (1983). Growth, development, and aging of human vocal folds. In D.M. Bless & J.H. Abbs (Eds.), Vocal Fold Physiology (pp. 22-43). San Diego, CA: College- Hill. Hirano, M., Takeuchi, Y., & Hiroto, I. (1966). Intranasal sound pressure during utterance of speech sounds. Folia Phoniatrica et Logopaedica, 18(5), 369–381. doi: 10.1159/000263113. Hixon, T. J., Weismer, G., & Hoit, J. D. (2008). Preclinical speech science: Anatomy, physiology, acoustics, and perception. San Diego, CA: Plural Publishing.

125

Hoch, L., Golding-Kushner, K., Siegel-Sadewitz, V.L., and Shprintzen, R.J. (1986). Speech Therapy. Seminars in Speech and Language, 7(3), 313-326. Hopkins, C. (2016). Nose, nasal cavity and paranasal sinuses. In S. Standring (Ed.), Gray's Anatomy: The Anatomical Basis of Clinical Practice (41st ed., pp. 556-570). New York, NY: Elsevier. Horiguchi, S., & Bell-Berti, F. (1987). The velotrace: A device for monitoring velar position. The Cleft Palate Journal, 24(2), 104-111. Horii, Y. (1980). An accelerometric approach to nasality measurement: a preliminary report. Cleft Palate Journal, 17, 254–261. Houde, J. F., & Jordan, M. I. (1998). Sensorimotor adaptation in speech production. Science, 279, 1213- 1216. Houde, J. F., & Nagarajan, S. S. (2011). Speech production as state feedback control. Front Hum Neurosci, 5(82), 1-14. doi: 10.3389/fnhum.2011.00082 House, A.S., & Stevens, K.N. (1956). Analog studies of the nasalization of vowels. Journal of Speech and Hearing Disorders, 21(2), 218-232. doi: 10.1044/jshd.2102.218 Huckel, W. (1845). Practical instructions in the art of singing. London, UK: H. Falkner. Hunt, O., Burden, D., Hepper, P., & Johnston, C. (2005). The psychosocial effects of cleft lip and palate: a systematic review. European Journal of Orthodontics, 27(3), 274-285. doi: 10.1093/ejo/cji004. Hurtado, V. (2005). Three approaches for teaching laryngeal stability: curing the 'necktie tenor' (Unpublished doctoral dissertation). Florida State University, Tallahassee, Florida. Husler, F., & Rodd-Marling, Y. (1976). Singing: The physical nature of the vocal organ: a guide to the unlocking of the singing voice. London, UK: Hutchinson. Idle, M., & Monaghan, A. (2016). Challenging concepts in oral and maxillofacial surgery: Cases with expert commentary. Oxford, UK: Oxford University Press. Iwarsson, J., & Sundberg, J. (1998). Effects of lung volume on vertical larynx position during phonation. Journal of Voice, 12(2), 159–165. doi: 10.1016/S0892-1997(98)80035-0 Jennings, J. J., & Kuehn, D. P. (2008). The effects of frequency range, vowel, dynamic loudness level, and gender on nasalance in amateur and classically trained singers. Journal of Voice, 22(1), 75- 89. doi:10.1016/j.jvoice.2006.08.017 John, A., Sell, D., Sweeney, T., Harding-Bell, A., & Williams, A. (2006). The Cleft Audit Protocol for Speech—Augmented: A Validated and Reliable Measure for Auditing Cleft Speech. The Cleft Palate-Craniofacial Journal, 43(3), 272–288. doi: 10.1597/04-141.1

126

Johnstone-Douglas, W. (1989). The teaching of Jean de Reszke. In B. Coffin (Ed.), Historical vocal pedagogy classics. Lanham, Md: Scarecrow Press. Jones, C., Schellenberg, M., & Gick, B. (2017). Indie-pop voice: How a pharyngeal/retracted articulatory setting may be driving a new singing style. Canadian Acoustics - Acoustique Canadienne, 45(3), 180-181. Jones, J. A., & Keough, D. (2008). Auditory-motor mapping for pitch control in singers and non- singers. Experimental Brain Research, 190(3), 279-287. doi:10.1007/s00221-008-1473-y Jones, J. A., & Munhall, K. G. (2003). Learning to produce speech with an altered vocal tract: The role of auditory feedback. The Journal of the Acoustical Society of America, 113(1), 532-543. doi: 10.1121/1.1529670 Jones, M.C. (1988). Etiology of facial clefts: prospective evaluation of 428 patients. Cleft Palate Journal 25, 16–20. Jordan, H. N., Schenck, G. C., Ellis, C., Rangarathnam, B., Fang, X., & Perry, J. L. (2017). Examining velopharyngeal closure patterns based on anatomic variables. Journal of Craniofacial Surgery, 28(1), 270-274. doi:10.1097/SCS.0000000000003284 Karnell, M.P. (2011). Instrumental assessment of velopharyngeal closure for speech. Seminars in Speech and Language, 32(2), 168-178. doi: 10.1055/s-0031-1277719 Kataoka, R., Michi, K.-I., Okabe, K., Miura, T., & Yoshida, H. (1996). Spectral Properties and Quantitative Evaluation of Hypernasality in Vowels. The Cleft Palate-Craniofacial Journal, 33(1), 43–50. doi: 10.1597/1545-1569_1996_033_0043_spaqeo_2.3.co_2 Kataoka, R., Warren, D. W., Zajac, D. J., Mayo, R., & Lutz, R. W. (2001). The relationship between spectral characteristics and perceived hypernasality in children. The Journal of the Acoustical Society of America, 109(5), 2181-2189. doi: 10.1121/1.1360717 Kato, K., Ueno, K., & Kawai, K. (2015). Effect of room acoustics on musicians' performance, Part II: Audio analysis of the variations in performed sound signals. Acta Acust. Acust 101, 743-759. Kawahara, H., & Williams, J. C. (1996). Effects of auditory feedback on voice pitch trajectories: Characteristic responses to pitch perturbations. In P.J. Davis & N.H. Fletcher (Eds.), Vocal fold physiology: controlling complexity and chaos (pp. 263–278). Sydney, AUS: Singular. Kent, R. D. (1996). Hearing and believing: Some limits to the auditory-perceptual assessment of speech and voice disorders. American Journal of Speech-Language Pathology, 5(3), 7-23. Keuning, K. H., Wieneke, G. H., & Dejonckere, P. H. (2004). Correlation between the perceptual rating of speech in Dutch patients with velopharyngeal insufficiency and composite measures derived

127

from mean nasalance scores. Folia Phoniatrica et Logopaedica, 56, 157–164. doi: 10.1159/000076937 Keuning, K.H., Wieneke, G.H., Van Wijngaarden, H.A., & Dejonckere, P.H. (2002). The correlation between nasalance and a differentiated perceptual rating of speech in Dutch patients with velopharyngeal insufficiency. The Cleft Palate-Craniofacial Journal, 39(3), 277-284. doi: 10.1597/1545-1569_2002_039_0277_tcbnaa_2.0.co_2. Kirkpatrick, A. (2009). Chiaroscuro and the quest for optimal resonance. Journal of Singing, 66(1), 15- 21. Kuehn, D. P. (1991). New therapy for treating hypernasal speech using continuous positive airway pressure (CPAP). Plastic and Reconstructive Surgery, 88(6), 959–969. Kuehn, D. P., & Moller, K. T. (2000). Speech and language Issues in the Cleft Palate population: The state of the art. The Cleft Palate-Craniofacial Journal, 37(4), 1–35. doi: 10.1597/1545- 1569_2000_037_0348_saliit_2.3.co_2 Kuehn, D. P., & Moon, J. B. (1998). Velopharyngeal closure force and levator veli palatini activation levels in varying phonetic contexts. Journal of Speech Language and Hearing Research, 41(1), 51-62. Kuehn, D. P., & Perry, J. L. (2009). Anatomy and physiology of the velopharynx. In J.E. Losee & R.E. Kirschner (Eds.), Comprehensive Cleft Care (pp. 557-68). New York, NY: McGraw Hill Medical. Kuehn, D. P., Imrey, P. B., Tomes, L., Jones, D. L., O'Gara, M. M., Seaver, E. J., Smith, B.E., Van Demark, D.R., & Wachtel, J. M. (2002). Efficacy of continuous positive airway pressure for treatment of hypernasality. The Cleft Palate-Craniofacial Journal, 39, 267-276. doi: 10.1597/1545-1569(2002)039<0267:EOCPAP>2.0.CO;2 Kuehn, D.P., & Moon, J.B. (2005). Histologic study of intravelar structures in normal human adult specimens. The Cleft Palate-Craniofacial Journal, 42(5), 481-489. doi: 10.1597/04-125r.1 Kummer, A.W. (2005). Simplified nasometric assessment procedures (SNAP): Nasometer test - revised. Lincoln Park, NJ: Kay Elemetrics. Kummer, A. (2011). Disorders of resonance and airflow secondary to cleft palate and/or velopharyngeal dysfunction. Seminars in Speech and Language, 32(2), 141-149. doi: 10.1055/s-0031-1277716 Kummer, A.W. (2011). Speech therapy for errors secondary to cleft palate and velopharyngeal dysfunction. Seminars in Speech and Language, 32(2), 191-198. doi: 10.1055/s-0031-1277721.

128

Kummer, A.W. (2014). Cleft palate and craniofacial anomalies – Effects on speech and resonance (3rd ed.). New York, NY: Delmar Cengage Learning. Kummer, A.W., Clark, S. L., Redle, E. E., Thomsen, L. L., & Billmire, D. A. (2012). Current practice in assessing and reporting speech outcomes of cleft palate and velopharyngeal surgery: a survey of cleft palate/craniofacial professionals. The Cleft Palate-Craniofacial Journal, 49, 146-152. doi: 10.1597/10-285 Kummer, A.W., & Lee, L. (1996). Evaluation and treatment of resonance disorders. Language, Speech, and Hearing Services in Schools, 27(3), 271-281. doi: 10.1044/0161-1461.2703.271 Lamperti, G. (1905). The Technics of bel canto. Translated by Theodore Baker. New York, NY: G. Schirmer. Lane, H., & Tranel, B. (1971). The Lombard sign and the role of hearing in speech. Journal of Speech and Hearing Research, 14, 677-709. doi: 10.1044/jshr.1404.677 Larson, C. R., Burnett, T. A., Kiran, S., & Hain, T. C. (2000). Effects of pitch-shift velocity on voice F0 responses. Journal of the Acoustical Society of America, 107(1), 559-564. doi: 10.1121/1.428323 Lass, N.J., & Pannbacker, M. (2015). Low-Tech Assessment of Resonance Disorders. Evidence-Based Communication Assessment and Intervention, 9(1), 43–50. doi: 10.1080/17489539.2015.1062258 Lawrence, V. (1979). Laryngological observations on belting. Journal of Research in Singing, 2(1), 26– 28. LeBorgne, W. D. L., & Rosenberg, M. D. (2014). The vocal athlete. San Diego, CA: Plural. Lee, A. S. Y., Ciocca, V., & Whitehill, T. L. (2004). Spectral analysis of hypernasality. Journal of Medical Speech-Language Pathology, 12(4), 173-177. Lee, A., Gibbon, F. E., & Spivey, K. (2017). Children’s attitudes toward peers with unintelligible speech associated with cleft lip and/or palate. Cleft Palate Craniofac J, 54(3), 262-268. doi: 10.1597/15-088 Leeper, H.A., Charles, D.H., & Sills, P.S. (2009). Prosthodontic management of maxillofacial and palatal defects. In K.T. Moller & L.E. Glaze (Eds.), Cleft palate: interdisciplinary issues and treatment (2nd ed., pp. 145-188). Austin, TX: Pro-Ed. Lehmann, L., Aldrich, R., & Willenbücher, C. (1927). How to sing (Meine Gesangskunst). New York, NY: Macmillan.

129

Levin, F. R. (2009). Greek reflections on the nature of music. Cambridge, UK: Cambridge University Press. Lewis, J.R., Andreassen, M.L., Leeper, H.A., Macrae, D.L., & Thomas, J. (1993). Vocal characteristics of children with cleft-lip/palate and associated velopharyngeal incompetence. Journal of Otolaryngology 22(2), 113-117. Lewis, K. E., & Watterson, T. (2003). Comparison of Nasalance Scores Obtained from the Nasometer and the NasalView. The Cleft Palate-Craniofacial Journal, 40(1), 40–45. https://doi.org/10.1597/1545-1569_2003_040_0040_consof_2.0.co_2 Lewis, K. E., Watterson, T., & Quint, T. (2000). The effect of vowels on nasalance scores. The Cleft Palate-Craniofacial Journal, 37, 584-589. doi: 10.1597/1545- 1569(2000)037<0584:TEOVON>2.0.CO;2 Lewis, K.E., Watterson, T., & Blanton, A. (2008). Comparison of short-term and long-term variability in nasalance scores. The Cleft Palate-Craniofacial Journal, 45, 495-500. doi: 10.1597/07- 150.1. Lindsay, R.B (1973). Acoustics: historical and philosophical development. Stroudsburg, PA: Dowden, Hutchinson & Ross. Lindsay, R.B. (1945). Historical development of acoustics to the time of Rayleigh. In J.W.S. Rayleigh (Ed.), The theory of sound (pp. xi-xxv). New York, NY: Dover Publications. Litzaw, L. L., & Dalston, R. M. (1992). The effect of gender upon nasalance scores among normal adult speakers. Journal of Communication Disorders, 25(1), 55-64. doi: 10.1016/0021- 9924(92)90014-N. Lockhart, E. (2003). The mental health needs of children and adolescents with cleft lip and/or palate. Clinical Child Psychology and Psychiatry, 8(1), 7-16. doi: 10.1177/1359104503008001003. Lof, G. (2008). Introduction to controversies about the use of nonspeech oral motor exercises. Seminars in Speech and Language, 29(4), 253-256. doi: 10.1055/s-0028-1103388 Logjes, R., Bleys, R., & Breugem, C. (2016). The innervation of the soft palate muscles involved in cleft palate: A review of the literature. Clinical Oral Investigations, 20(5), 895-901. doi:10.1007/s00784-016-1791-6 Lohmander, A., Persson, C., Willadsen, E., Lundeborg, I., Alaluusua, S., Aukner, R., Bau, A., Boers, M., Bowden, M., Davies, J., Emborg, B., Havstam, C, Hayden, C., Henningsson, G., Holmefjord, A., Hölttä, C., Kisling-Møller, M., Kjøll, L., Lundberg, M., McAleer, E., Nyberg, J., Paaso, M., Pedersen, N.H., Rasmussen, T., Reisæter, S., Søgaard Andersen, H., Schöps, A.,

130

Tørdal, I-B., & Semb, G. (2017). Scandcleft randomised trials of primary surgery for unilateral cleft lip and palate: 4. Speech outcomes in 5-year-olds - velopharyngeal competency and hypernasality. J Plast Surg Hand Surg, 51(1), 27-37. doi:10.1080/2000656X.2016.1254645 Lohmander, A., Willadsen, E., Persson, C., Henningsson, G., Bowden, M., & Hutters, B. (2009). Methodology for Speech Assessment in the Scandcleft Project—An International Randomized Clinical Trial on Palatal Surgery: Experiences from a Pilot Study. The Cleft Palate-Craniofacial Journal, 46(4), 347–362. doi: 10.1597/08-039.1 Lombard, E. (1911). Le signe de l'elevation de la voix. Annales des Maladies de l' Oreille, du Larynx, du Nez et du. Pharynx, 37, 101-119. Lovetri, J. L. (2002). Contemporary commercial music: More than one way to use the vocal tract. Journal of Singing - the Official Journal of the National Association of Teachers of Singing, 58(3), 249-252. Lovetri, J., Lesh, S., & Woo, P. (1999). Preliminary study on the ability of trained singers to control the intrinsic and extrinsic laryngeal musculature. Journal of Voice, 13(2), 219-226. doi:10.1016/S0892-1997(99)80024-1 Lubker, J. (1968). An electromyographic-cinefluorographic investigation of velar function during normal speech production. Cleft Palate Journal, 5(1), 1–17 Lubker, J.F. & Moll, K.L (1965). Simultaneous oral-nasal air flow measurements and cinefluorographic observations during speech production. The Cleft Palate Journal, 36, 257-272. Luizard, P., Brauer, E., Weinzierl, S., & Henrich Bernardoni, N. (2018). How singers adapt to room acoustical conditions. Proceedings of the Institute of Acoustics. Institute of Acoustics, 40(3). doi: 10.14279/depositonce-8704 Luizard, P., Steffens, J., & Weinzierl, S. (2020). Singing in different rooms: Common or individual adaptation patterns to the acoustic conditions? Journal of the Acoustical Society of America, 147(2), EL132-EL137. MacKay, I.R.A., & Kummer, A. W. (1994). Simplified nasometric assessment procedures. In Kay Elemetrics Corp. (Ed.), Instruction manual: Nasometer model 6200-3 (pp. 123–142). Lincoln Park, NJ: Kay Elemetrics Corp. Maeda, S. (1990). Compensatory articulation during speech: Evidence from the analysis and synthesis of vocal tract shapes using an articulatory model. In W.J. Hardcastle & A. Marchal (Eds.), Speech production and speech modeling (pp. 131-149). Boston, MA: Kluwer Academic Publishers.

131

Mainka, A., Poznyakovskiy, A., Platzek, I., Fleischer, M., Sundberg, J., & Mürbe, D. (2015). Lower vocal tract morphologic adjustments are relevant for voice timbre in singing. PLoS One, 10(7), e0132241. doi:10.1371/journal.pone.0132241 Malde, M. (2009). Mapping the structures of resonance. Journal of Singing - the Official Journal of the National Association of Teachers of Singing, 65(5), 521-529. Mannstein, H.F. (1848). Die Grosse Italienische Gesangschule. Dresden, DE: Arnoldischen Buchhandlung. Marafioti, P.M. (1922). Caruso's method of voice production: The scientific culture of the voice. New York, NY: Appleton. Marchesi, M. (1898). Marchesi and music: passages from the life of a famous singing teacher. New York, NY; London, UK: Harper & Bros. Publishers. Mason, D. (2000). The teaching (and learning) of singing. In J. Potter (Ed.), The Cambridge companion to singing (pp. 204-220). Cambridge, UK: Cambridge University Press. doi:10.1017/CCOL9780521622257.018 Matsuya, T., Yamaoka, M., & Miyasaki, T. (1979). A fiberoptic study of velopharyngeal closure in patients with operated cleft palates. Plast Reconstr Surg, 63, 497-500. Mayo, R., Warren, D. W., & Zajac, D. J. (1998). Intraoral Pressure and Velopharyngeal Function. The Cleft Palate-Craniofacial Journal, 35(4), 299–303. doi: 10.1597/1545- 1569_1998_035_0299_ipavf_2.3.co_2 McCauley, R., Strand, E., Lof, G.L., Schooling, T., & Frymark, T. (2009). Evidence-based systematic review: Effects of non-speech oral motor exercises on speech. American Journal of Speech- Language Pathology, 18, 343-360. doi: 10.1044/1058-0360(2009/09-0006) McCoy, S. (2004). Your voice: an inside view. Delaware, OH: Inside View Press. McCoy, S. (2008). The seduction of nasality. Journal of Singing, 64(5), 579-582. McDonald Klimek, M. (2008). Using twang. In A. Behrman & J. Haskell (Eds.), Exercises for voice therapy. San Diego, CA: Plural Publishing. McDonald, E.T., & Koepp-Baker, H. (1951). Cleft palate speech: an integration of research and clinical observation. Journal of Speech and Hearing Disorders, 16, 9-20. doi: 10.1044/jshd.1601.09. McHenry, M.A. (1997). The effect of increased vocal effort on estimated velopharyngeal orifice area. American Journal of Speech-Language Pathology, 6(4), 55-61. doi: 10.1044/1058-0360.0604.55. McIver, W., & Miller, R. (1996). A brief study of nasality in singing. Journal of Singing - the Official Journal of the National Association of Teachers of Singing, 52(4), 21-26.

132

McKinney, J.C. (2005). The diagnosis and correction of vocal faults. (Revised ed.). Longrove, IL: Waveland Press Inc. McWilliams, B. J., & Bradley, D. P. (1965). Ratings of velopharyngeal closure during blowing and speech. Cleft Palate Journal, 45, 46–55. McWilliams, B. J., Morris, H. L., & Shelton, R. L. (1990). Cleft palate speech (2nd ed.). Philadelphia, PA: B.C. Decker. McWilliams, B.J., & Phillips, B.J. (1979). Velopharyngeal incompetence: Audio seminars in speech pathology. Philadelphia, PA: W.B. Saunders, Inc. Mersenne, M., & Chapman, R. E. (1957). Harmonie universelle: the books on instruments. The Hague: M. Nijhoff. Miller, D.G. (2008). Resonance in singing: Voice building through acoustic feedback. Princeton, NJ: Inside View Press. Miller, R. (1977). English, French, German and Italian techniques of singing. Metuchen, NJ: Scarecrow Press. Miller, R. (1983). Care of the professional voice: Transcripts of the eleventh symposium care of the professional voice. New York, NY: The Voice Foundation. Miller, R. (1986). The structure of singing: System and art in vocal technique. New York, NY: Schirmer. Miller, R. (1993). Training tenor Voices. New York, NY: Schirmer Books. Miller, R. (1996). The velopharyngeal (palatopharyngeal) port during singing. Journal of Singing - the Official Journal of the National Association of Teachers of Singing, 53(1), 27-29 Miller, R. (1997). National schools of singing: English, French, German, and Italian techniques of singing revisited. Lanham, MD: Scarecrow Press. Miller, R. (2004). Solutions for singers: Tools for performers and teachers. Oxford, UK: Oxford University Press. Mitchell, L. E., Beaty, T. H., Lidral, A. C., Munger, R. G., Murray, J. C., Saal, H. M., & Wyszynski, D. F. (2002). Guidelines for the Design and Analysis of Studies on Non-syndromic Cleft Lip and Cleft Palate in Humans: Summary Report from a Workshop of the International Consortium for Oral Clefts Genetics. The Cleft Palate-Craniofacial Journal, 39(1), 93–100. doi: 10.1597/1545- 1569_2002_039_0093_gftdaa_2.0.co_2 Moll, K. L. (1962). Velopharyngeal closure on vowels. Journal of Speech, Language and Hearing Research, 5(1), 30-37. doi: 10.1044/jshr.0501.30

133

Moon J.B., & Kuehn D.P. (2004). Anatomy and physiology of normal and disordered velopharyngeal function for speech. In K. R. Bzoch (Ed.), Communicative disorders related to cleft lip and palate (5th ed., pp. 67-98). Austin, TX: Pro-Ed. Moon, J. B. (2009). Evaluation of velopharyngeal function. In K.T. Moller & L.E. Glaze (Eds.), Cleft lip and palate: Interdisciplinary issues and treatment (2nd ed., pp. 313-76). Austin TX: Pro-ed Publishing. Moon, J. B., & Kuehn, D. P. (1996): Anatomy and physiology of normal and disordered velopharyngeal function for speech. National Center for Voice and Speech, 9, 143–158. Moon, J.B., Kuehn, D.P., & Huisman, J.J. (1994). Measurement of velopharyngeal closure force during vowel production. The Cleft Palate Craniofacial Journal, 31(5), 356-363. doi: 10.1597/1545- 1569_1994_031_0356_movcfd_2.3.co_2. Moore, K. L., & Persaud, T. V. N. (1998). The developing human: Clinically oriented embryology. Philadelphia, PA: Saunders. Moore, W.H., & Sommers, R.K. (1973). Phonetic contexts: Their effects on perceived nasality in cleft palate speakers. The Cleft Palate Journal, 10(1), 72-83. Morley, M.E. (1970). Cleft palate and speech. Edinburgh, UK: Livingstone. Morris, H.L. (1972). Cleft palate. In A.J. Watterson (Ed.), Communicative disorders. Springfield, IL: Thomas. Mossey, P.A., & Modell, B. (2012). Epidemiology of oral clefts 2012: An international perspective. Front Oral Biol 16, 1-18. Müller, J. (1848). The physiology of the sense, voice and muscular motion with the mental faculties. Translator: W. Baly. London, UK: Walton & Maberly. Munhall, K. G., MacDonald, E. N., Byrne, S. K., & Johnsrude, I. (2009). Talkers alter vowel production in response to real-time formant perturbation even when instructed not to compensate. The Journal of the Acoustical Society of America, 125, 384-390. doi:10.1121/1.3035829 Nagarajan, R., Savitha, V. H., & Subramaniyan, B. B. (2009). Communication disorders in individuals with cleft lip and palate: An overview. Indian Journal of Plastic Surgery, 42, 137-143. doi: 10.4103/0970-0358.57199 Nagel, W. (1909). Physiologie der Stimmwerkzeuge. In W. Nagel (Ed.), Handbuch der Physiologie des Mensche (pp. 691-792). Braunschweig, DE: Friedrich Vieweg und Sohn. Nair, A., Nair, G., & Reishofer, G. (2016). The low mandible maneuver and its resonential implications for elite singers. Journal of Voice, 30(1), 128.e13-32. doi:10.1016/j.jvoice.2015.03.010

134

Nair, G. (2007). The craft of singing. San Diego, CA: Plural Publishing. Nix, J. (1999). Lip trills and raspberries: "high spit factor": Alternatives to the nasal continuant consonants. Journal of Singing - the Official Journal of the National Association of Teachers of Singing, 55(3), 15-19. Nye, E. (2014). Chiaroscuro in the Eighteenth‐Century Ballet d'Action. Journal of 18th-Century Studies, 37, 487-498. doi:10.1111/1754-0208.12202 Paal, S., Reulbach, U., Strobel-Schwarthoff, K., Nkenke, E., & Schuster. M. (2005). Evaluation of speech disorders in children with cleft lip and palate. Journal of Orofacial Orthopedics, 66(4), 270–278. doi: 0.1007/s00056-005-0427-2. Pabst, F., & Sundberg, J. (1993). Tracking multi-channel electroglottograph measurement of larynx height in singers. Scandinavian Journal of Logopedics and Phoniatrics, 18(4), 143–152. doi: 10.3109/14015439309101360 Parrell, B., Lammert, A.C., Ciccarelli, G., & Quatieri, T.F. (2019). Current models of speech motor control: A control-theoretic overview of architectures and properties. The Journal of the Acoustical Society of America 145(3), 1456-1481. doi: 10.1121/1.5092807 Patel, A., Clune, J. E., Steinbacher, D. M., & Persing, J. A. (2013). Comprehensive cleft center: a paradigm shift in cleft care. Plast Reconstr Surg, 131(2), 312e-313e. Doi: 10.1097/PRS.0b013e31827a294a Patri, J. F., Perrier, P., Schwartz, J. L., & Diard, J. (2018). What drives the perceptual change resulting from speech motor adaptation? Evaluation of hypotheses in a Bayesian modeling framework. PLoS Comput Biol, 14(1), e1005942. doi:10.1371/journal.pcbi.1005942 Perkell, J.S. (2012). Movement goals and feedback and feedforward control mechanisms in speech production. Journal of Neurolinguistics, 25(5), 382-407. doi: 10.1016/j.jneuroling.2010.02.011 Perkell, J. S., Guenther, F. H., Lane, H., Matthies, M. L., Perrier, P., Vick, J., Wilhelms-Tricarico, R., & Zandipour, M. (2000). A theory of speech motor control and supporting data from speakers with normal hearing and with profound hearing loss. Journal of Phonetics, 28(3), 233-272. doi:10.1006/jpho.2000.0116 Perkins, W.H. (1983). Current therapy of communication disorders: dysarthria and apraxia. New York, NY: Thieme-Stratton. Perry, J. L. (2011). Anatomy and physiology of the velopharyngeal mechanism. Seminars in Speech and Language, 32(2), 83-92. doi: 10.1055/s-0031-1277712

135

Perry, J. L., Sutton, B.P., Kuehn, D.P., & Gamage, J.K. (2014). Using MRI for assessing velopharyngeal structures and function. The Cleft Palate - Craniofacial Journal, 51(4), 476-85. doi: 10.1597/12- 083. Perry, J., & Zajac, D. J. (2017). Orofacial and velopharyngeal structure and function. In D.J. Zajac & L.D. Vallino (Eds.), Evaluation and management of cleft lip and palate: A developmental perspective (pp. 3-21). San Diego, CA: Plural Publishing. Perry, J.L., Mason, K., Sutton, B.P., & Kuehn, D.P. (2018). Can dynamic MRI be used to accurately identify velopharyngeal closure patterns? The Cleft Palate-Craniofacial Journal, 55(4), 499– 507. doi: 10.1177/1055665617735998 Peter, S., Abdul Rahman, Z.A., & Pillai, S. (2019). Hypernasality in singing among children with cleft palate: a preliminary study. International Journal of Oral and Maxillofacial Surgery. 48(10), 1317-1322. https://doi.org/10.1016/j.ijom.2019.03.896. Peterson-Falzone, S. J., Hardin-Jones, M. A., & Karnell, M. P. (2001). Cleft palate speech (3rd ed.). St Louis, MO: Mosby, Inc. Peterson-Falzone, S. J., Hardin-Jones, M. A., & Karnell, M. P. (2010). Cleft palate speech (4th ed.). St. Louis, Mo: Mosby/Elsevier. Peterson-Falzone, S. J., Trost-Cardamone, J. E., Karnell, M. P., & Hardin-Jones M. A. (2006). The clinician's guide to treating cleft palate speech. St. Louis, MO: Mosby-Elsevier. Picheny, M. A., Durlach, N. I., & Braida, L. D. (1989). Speaking clearly for the hard of hearing. III: An attempt to determine the contribution of speaking rate to differences in intelligibility between clear and conversational speech. J Speech Hear Res, 32(3), 600-603. doi: 10.1044/jshr.3203.600 Poiseuille, J.L.M. (1846). Recherches experimentales sur Ie mouvement des liquides dans les tubes de tres-petits diametres. In Memoires presentes par divers savants a i'Academie Royale des Sciences de l'Institut de France, IX, 433-544 Purcell, D. W., & Munhall, K. G. (2006). Compensation following real-time manipulation of formants in isolated vowels. Journal of the Acoustical Society of America, 119(4), 2288-2297. doi: 10.1121/1.2173514 Ramanarayanan, V., Van Segbroeck, M., & Narayanan, S.S. (2016). Directly data-derived articulatory gesture-like representations retain discriminatory information about phone categories. Computer Speech & Language, 36, 330-346. doi:10.1016/j.csl.2015.03.004 Reid, C. (1990). The nature of resonance. Journal of Research in Singing and Applied Vocal Pedagogy, 14(1), 1-26.

136

Reisberg, D. J. (2000). Dental and Prosthodontic Care for Patients with Cleft or Craniofacial Conditions. The Cleft Palate-Craniofacial Journal, 37(6), 534–537. doi: 10.1597/1545- 1569_2000_037_0534_dapcfp_2.0.co_2 Richman, L.C., McCoy, T.E., Conrad, A.L., & Nopoulos, P.C. (2012). Neuropsychological, behavioral, and academic sequelae of cleft: early developmental, school age, and adolescent/young adult outcomes. The Cleft Palate Craniofacial Journal, 49(4), 387-396. doi: 10.1597/10-237. Riemann, B.L., & Lephart, S.M. (2002). The sensorimotor system, part II: the role of proprioception in motor control and functional joint stability. Journal of Athletic Training 37(1), 80-84. Roberts, G. E. (2016). From music to mathematics: Exploring the connections. Baltimore, MD: Johns Hopkins University Press. Rong, P., & Kuehn, D. (2012). The effect of articulatory adjustment on reducing hypernasality, Journal of Speech. Language and Hearing Research, 55(5), 1438-1448. doi: 10.1044/1092- 4388(2012/11-0142). Rossing, T. D. (2007). Springer handbook of acoustics. New York, NY: Springer. Rua Ventura, S.M., Freitas, D.R.S., Ramos, I.M.A.P., & Tavares, J.M.R.S. (2013). Morphologic differences in the vocal tract resonance cavities of voice professionals: An MRI-based study. Journal of Voice, 27(2), 132–140. doi: 10.1016/j.jvoice.2012.11.010 Ruscello, D. (2008a). Non-speech oral motor treatment issues related to children with developmental speech sound disorders. Language, Speech and Hearing Services in Schools, 39, 380-391. doi: 10.1044/0161-1461(2008/036) Ruscello, D. (2008b). An examination of non-speech oral motor exercises for children with velopharyngeal inadequacy. Seminars in Speech and Language, 29, 294-303. doi: 10.1055/s- 0028-1103393 Ruscello, D.M., Shuster, L.J., & Sandwisch, A. (1991). Modification of context-specific nasal emission. Journal of Speech and Hearing Research, 34, 27–32. doi: 10.1044/jshr.3401.27. Russell G.O. (1931). Speech and voice. New York, NY: MacMillan. Saltzman, E. L., & Munhall, K. G. (1989). A dynamical approach to gestural patterning in speech production. Ecological Psychology, 1(4), 333-382. doi:10.1207/s15326969eco0104_2 Saltzman, E., & Kelso, J.A.S. (1987). Skilled actions: A task dynamic approach. Psychological Review, 94(1), 84-106. doi: 10.1037/0033-295X.94.1.84

137

Santoni, C., de Boer, G., Thaut, M., & Bressmann, T. (2018). Influence of altered auditory feedback on oral-nasal balance in song. Journal of Voice, 34(1), 157.e9-157.e15. doi:10.1016/j.jvoice.2018.06.014 Santoni, C., de Boer, G., Thaut, M., & Bressmann, T. (2019, in press). Influence of Voice Focus Adjustments on Oral-Nasal Balance in Speech and Song. Folia Phoniatr Logop. doi: 10.1159/000501908 Santoni, C., Thaut, M., & Bressmann, T. (2020, in press). Immediate effects of voice focus adjustments on hypernasal speakers’ nasalance scores. International Journal of Pediatric Otorhinolaryngology. Sataloff, R. (1997). Professional Voice: the science and art of clinical care (2nd ed.). San Diego, CA: Singular Publishing Group. Sataloff, R.T., & Titze, I.R. (1991). Vocal health and science. Jacksonville, FL: The National Association of Teachers of Singing. Scearce, L. (2016). Manual of singing voice rehabilitation: A practice approach to vocal health and wellness. San Diego, CA: Plural Publishing, Inc. Scheerer, N. E., & Jones, J. A. (2012). The relationship between vocal accuracy and variability to the level of compensation to altered auditory feedback. Neuroscience Letters, 529(2), 128-132. doi:10.1016/j.neulet.2012.09.012 Schmitt, F. (1854). Grosse Gesangschule für Deutschland. München: Beim Verfasser. Schoen Rene, A.E. (1941). America’s musical inheritance – Memories and reminiscences. New York, NY: G.P Putnam’s Sons. Schroeder, M. R. (1993). A brief history of synthetic speech. Speech Communication, 13(1-2), 231-237. doi: 10.1016/0167-6393(93)90074-U Schutte, H. K., & Miller, D. G. (1988). Resonance spectrum of the singing voice relative to supra- and sub-glottal pressure changes; consequences for the theory of voice production. Folia Phoniatrica et Logopaedica 40(2), 65-73. Schutte, H. K., & Miller, D. G. (1993). Belting and pop, nonclassical approaches to the female middle voice: Some preliminary considerations. Journal of Voice, 7(2), 142-150. doi:10.1016/S0892- 1997(05)80344-3 Scotto di Carlo, N. (1994). Internal voice sensitivities in opera singers. Folia Phoniatr Logop, 46(2), 79-85.

138

Scotto di Carlo, N. & Autessere, D. (1987). Movements of the velum in singing. Journal of Research in Singing and Applied Vocal Pedagogy, 11(1), 3-13. Seaver, E.J., Dalston, R.M., Leeper, H.A., & Adams, L.E. (1991). A study of nasometric values for normal nasal resonance. Journal of Speech, Language, and Hearing Research, 34, 715-722. doi: 10.1044/jshr.3404.715. Sedlackova, E., Lastovka, M., Sram, F. (1973). Contribution to knowledge of soft palate innervation. Folia Phoniatrica 25, 434-441. doi: 10.1159/000263731 Sell, D., Harding, A., & Grunwell, P. (1999). GOS.SP.ASS.'98: An assessment for speech disorders associated with cleft palate and/or velopharyngeal dysfunction (revised). International Journal of Language & Communication Disorders, 34(1), 17-33. doi: 10.1080/136828299247595 Sell, D., John, A., Harding‐Bell, A., Sweeney, T., Hegarty, F., & Freeman, J. (2009). Cleft Audit Protocol for Speech (CAPS‐A): a comprehensive training package for speech analysis. International Journal of Language & Communication Disorders, 44(4), 529–548. doi: 10.1080/13682820802196815 Semmens, R. (1981). An early eighteenth-century discussion of musical acoustics by Etienne Loulié: Canadian Journal of Music. Canadian University Music Review, 2, 177-206. Shah, S., & Garritano, F. G. (2015). Pediatric oral anatomy. Operative Techniques in Otolaryngology - Head and Neck Surgery, 26(1), 2-7. doi:10.1016/j.otot.2015.01.007 Shakespeare, W. (1910). The art of singing. London: Metzler & Co. [Reprinted London: J.B. Cramer, (n.d.)]. Shankland, W.E. (2001). The trigeminal nerve: part IV. The mandibular division. Cranio 19, 153-161. doi: 10.1080/08869634.2001.11746164 Sharp, P., Kelly, S., Main, A., & Manley, G. (1999). An instrument for the multiparameter assessment of speech. Medical Engineering & Physics, 21(9), 661–671. doi: 10.1016/S1350-4533(99)00091-0 Shelton, R. L., Knox, A. W., Elbert, M., & Johnson, T. S. (1970). Palate awareness and nonspeech voluntary palate movement. In J. F. Bosma (Ed.), Second symposium on oral sensation and perception (pp. 416-431). Springfield, IL: Charles C. Thomas Publishing. Shelton, R.I., Hahn, E., & Morris, H. (1968). Diagnosis and therapy. In D.C. Spriestersbach & D. Sherman (Eds.), Cleft Palate and Communication (pp. 225-268). New York, NY: Academic Press.

139

Shiller, D. M., Sato, M., Gracco, V. L., & Baum, S. R. (2009). Perceptual recalibration of speech sounds following speech motor learning. Journal of the Acoustical Society of America, 125(2), 1103- 1113. doi:10.1121/1.3058638 Shimokawa, T., Yi, S., & Tanaka, S. (2005). Nerve supply to the soft palate muscles with special reference to the distribution of the lesser palatine nerve. The Cleft Palate-Craniofacial Journal, 42(5), 495-500. doi: 10.1597/04-142R.1 Shipp, T. (1984). Effects of vocal Frequency and effort on vertical laryngeal position. Journal of Research in Singing, 7(2), 1-5. Shipp, T. (1987). Vertical laryngeal position: Research findings and application for singers. Journal of Voice: Official Journal of the Voice Foundation, 1(3), 217-219. Shipp, T., & Izdebski, K. (1975). Letter: Vocal frequency and vertical larynx positioning by singers and nonsingers. Journal of the Acoustical Society of America, 58(5), 1104-1106. doi: 10.1121/1.380776 Shipp, T., & Morrissey, P. (1977). Physiologic adjustments for frequency change in trained and untrained voices. The Journal of the Acoustical Society of America 62(2), 476-479. Shipp, T., Guinn, L., Sundberg, J. & Titze, I.R. (1987). Vertical laryngeal position: research findings and their relationship to singing. Journal of Voice 1, 220-222. Shprintzen RJ, Bardach J. (1995). Cleft speech management, a multidisciplinary approach. St. Louis, MO: Mosby. Shprintzen, R. J., Lencione, R. M., McCall, G. N., & Skolnick, M. L. (1974). A three-dimensional cinefluoroscopic analysis of velopharyngeal closure during speech and nonspeech activities in normals. The Cleft Palate Journal, 11, 412-428. Shprintzen, R. J., Lewin, M. L., & Croft, C. B. (1979). A comprehensive study of pharyngeal flap surgery: tailor made flaps. The Cleft Palate Journal, 16(1), 46-55. Shprintzen, R. J., McCall, G. N., Skolnick, M. L., & Lencione, R. M. (1975). Selective movement of the lateral aspects of the pharyngeal walls during velopharyngeal closure for speech, blowing, and whilstling in normals. The Cleft Palate Journal, 12(1), 51-58. Shprintzen, R.J. (2013). The velopharyngeal mechanism. In S. Berkowitz (Ed.), Cleft lip and palate diagnosis and management (3rd ed., pp. 741-757). New York, NY: Springer. Siegel, G., & Pick, H. (1974). Auditory feedback in the regulation of voice. Journal of the Acoustical Society of America, 65, 1618-1624. doi: 10.1121/1.1903486

140

Simpson, A.P. (2009), Phonetic differences between male and female speech. Language and Linguistics Compass, 3, 621-640. doi:10.1111/j.1749-818X.2009.00125.x Skolnick, M.L. (1975). Velopharyngeal function in cleft palate. Clinics in plastic surgery, 2(2), 285- 297. Smith B., & Sataloff, R. (2013). Choral pedagogy (3rd ed.). San Diego, CA: Singular Publishing. Smith, B. E. & Kuehn, D. P. (2007). Speech evaluation of velopharyngeal dysfunction. Journal of Craniofacial Surgery, 18(2), 251-261. doi: 10.1097/SCS.0b013e31803ecf3b. Smith, W.S. (2007). The naked voice: A wholistic approach to singing. New York, NY: Oxford University Press. Sorenson, D.N. (1989). A fundamental frequency investigation of children ages 6-10 years old. Journal of Communication Disorders, 22(2), 115-123. doi: 10.1016/0021-9924(89)90028-2. Srinivas, V., & Bressmann, T. (2019). Effects of knowledge of task on control of oral-nasal balance in speech. Folia Phoniatrica et Logopaedica. doi: 10.1159/000503861 Stager, S. V., Bielamowicz, S. A., Regnell, J. R., Gupta, A., & Barkmeier, J. M. (2000). Supraglottic activity: Evidence of vocal hyperfunction or laryngeal articulation. Journal of Speech, Language, and Hearing Research, 43(1), 229-238. Stark, J. (1999). Bel canto: A history of vocal pedagogy. Toronto, ON: University of Toronto Press. Steinhauer, K & Estill, J. (2008). The Estill voice model: Physiology of emotion. In K. Izdebski (Ed.), Emotions of the human voice (Vol. 2, Clinical evidence, pp. 83-100). San Diego, CA: Plural Publishing. Stemple, J. C. (2000). Voice therapy: Clinical studies. Clifton Park, NY: Delmar Thomson Learning. Stemple, J.C., Glaze, L.E., & Klaben, B. (2010). Clinical voice pathology: Theory and management (4th ed.). San Diego, CA: Singular Publishing Group. Stock, N.M., & Feragen, K.B. (2018). Comparing psychological adjustment across cleft and other craniofacial conditions: implications for outcome measurement and intervention. The Cleft Palate Craniofacial Journal, 56(6), 766-772. doi: 10.1080/08870446.2016.1143944. Stockhausen, J. (1884). Method of singing. Translator: S. Löwe. London, UK: Novello. Stone Jr, R.E. (Ed), Cleveland, T.F., Sundberg, P.J., & Prokop, J. (2003). Aerodynamic and acoustical measures of speech, operatic, and broadway vocal styles in a professional female singer. The Journal of Voice, 17(3), 283–297. doi: 10.1067/S0892-1997(03)00074-2

141

Story, B. H., Titze, I. R., & Hoffman, E. A. (1996). Vocal tract area functions from magnetic resonance imaging. The Journal of the Acoustical Society of America, 100(1), 537-554. doi: 10.1121/1.415960 Story, B. H., Titze, I. R., & Hoffman, E. A. (2001). The relationship of vocal tract shape to three voice qualities. Journal of the Acoustical Society of America, 109(4), 1651-1667. doi: 10.1121/1.1352085 Sumida, K., Yamashita, K., & Kitamura, S. (2012). Gross anatomical study of the human palatopharyngeus muscles throughout its entire course from origin to insertion. Clinical Anatomy, 25(3), 314-323. Sundberg, J. (1970). Formant structure and articulation of spoken and sung vowels. Folia Phoniatr, 22, 28-48. Sundberg, J. (1974). Articulatory interpretation of the 'singing formant'. Journal of the Acoustical Society of America, 55, 838-844. Sundberg, J. (1987). The science of the singing voice. DeKalb, Il: Northern Illinois University Press. Sundberg, J. (1988). Vocal tract resonance in singing. The NATS Journal, March–April, 11–31. Sundberg, J. (1990). What’s so special about singers? Journal of Voice, 4(2), 107–119. doi: 10.1016/S0892-1997(05)80135-3 Sundberg, J. (1998). Vocal tract resonance. In R.T. Sataloff (Ed.), Vocal health and pedagogy (pp. 47- 65). San Diego, CA: Singular. Sundberg, J. & Nordström, P-E. (1976). Raised and lowered larynx – the effect on vowel formant Frequencies. Speech Transmission Laboratory Quarterly Progress and Status Reports, 17(2-3), 35-39. Sundberg, J., & Askenfelt, A. (1983). Larynx height and voice source: a relationship? In J. Abbs & D. Bless (Eds.). Vocal fold physiology (pp. 307-316). Houston, TX: College Hilll. Sundberg, J., & Thalén, M. (2010). What is 'twang'? Journal of Voice, 24(6), 654-660. doi: 10.1016/j.jvoice.2009.03.003. Sundberg, J., Birch, P., Gumoes, B., Stavad, H., Prytz, S., & Karle, A. (2007). Experimental findings on the nasal tract resonator in singing. Journal of Voice, 21(2), 127-137. doi:10.1016/j.jvoice.2005.11.005 Sundberg, J., Gramming, P., & Lovetri, J. (1993). Comparisons of pharynx, source, formant, and pressure characteristics in operatic and musical theatre singing. Journal of Voice, 7(4), 301- 310.doi: 10.1016/S0892-1997(05)80118-3

142

Sundberg, J., Leanderson, R., & von Euler, C. (1989). Activity relationship between diaphragm and cricothyroid muscles. Journal of Voice, 3(3), 225–232. doi: 10.1016/S0892-1997(89)80004-9 Sundberg, J., Thalén, M., & Popeil, L. (2012). Substyles of belting: Phonatory and resonatory characteristics. Journal of Voice, 26(1), 44-50. doi:10.1016/j.jvoice.2010.10.007 Sutera S.P., & Skalak, R. (1993). The history of Poiseuille's Law. Annu Rev Fluid Mech, 25(1), 1- 20. doi: 10.1146/annurev.fl.25.010193.000245. Suwaki M., Nanba K., Ito E., Kumakura I., & Minagi S. (2008). Nasal speaking valve: a device for managing velopharyngeal incompetence. Journal of Oral Rehabilitation, 35(1), 73-78. doi: 10.1111/j.1365-2842.2007.01800.x Svensson, E. (2000). Concordance between ratings using different scales for the same variable. Statistics in Medicine, 19(24), 3483–3496. doi: 10.1002/1097-0258(20001230)19:24<3483::AID- SIM786>3.0.CO;2-A Sweeney, T. (2011). Nasality: assessment and intervention. In S. Howard & A Lohmander (Eds.), Cleft palate speech: Assessment and intervention (pp. 199-220). West Sussex, UK: Wiley-Blackwell. Tanner, K., Roy, N., Merrill, R. M., & Power, D. (2005). Velopharyngeal port status during classical singing. J Speech Lang Hear Res, 48(6), 1311-1324. doi:10.1044/1092-4388(2005/091) Ternström, S. (1993). Long-time average spectrum characteristics of different choirs in different rooms. Voice (UK) 2, 55–77. Thaut, C. (2014). Vocal intonation therapy. In M.H. Thaut & V. Hoemberg (Eds.), The Oxford handbook of neurologic music therapy (pp. 179-184). Oxford, UK: Oxford University Press. Thaut, M.H. (2005). Neurologic music therapy in sensorimotor rehabilitation. In M.H. Thaut (Ed.), Rhythm, music and the brain (pp. 137-164). New York, NY: Routledge. Thaut, M.H., & Hoemberg, V. (2014). The Oxford handbook of neurologic music therapy. Oxford: Oxford University Press. Titze, I. R. (2001). Acoustic interpretation of resonant voice. The Journal of Voice, 15(4), 519-528. doi:10.1016/S0892-1997(01)00052-2 Titze, I. R. (2007). Belting and a high larynx position. Journal of Singing, 63(5), 557-558. Titze, I., et al. (1991). Vocal health and science. Jacksonville, FL: NATS. Titze, I.R. (1987). Nasality in vowels. The NATS Journal, 43(4), 34-37. Titze, I.R. (1994). Principles of voice production. Englewood Cliffs, NJ: Prentice Hall.

143

Titze, I.R., Baken, R.J., Bozeman, K.W., et al. (2015). Toward a consensus on symbolic notation of harmonics, resonances, and formants in vocalization. J Acoust Soc Am. 137(5), 3005-3007. doi:10.1121/1.4919349 Titze, I.R., Bergan, C. C., Hunter, E. J., & Story, B. (2003). Source and filter adjustments affecting the perception of the vocal qualities twang and yawn. Logopedics Phoniatrics Vocology, 28(4), 147- 155. Titze, I.R., Maxfield, L.M., & Walker, M.C. (2017). A Formant Range Profile for Singers. Journal of Voice, 31(3), 382.e9-382.e13. doi: 10.1016/j.jvoice.2016.08.014 Titze, I. R., & Story, B. H. (1997). Acoustic interactions of the voice source with the lower vocal tract. The Journal of the Acoustical Society of America, 101(4), 2234-2243. doi: 10.1121/1.418246 Titze, I.R., Worley, A. S., & Story, B. H. (2011). Source-vocal tract interaction in female operatic singing and theater belting. Journal of Singing, 67(5), 561-572. Tosi, F.P. (1743). Observations on the florid song; or, Sentiments on the ancient and modern singers (2nd ed.). Translated into English by J.E. Galliard. London, UK: J. Wilcox Tourville, J.A., & Guenther, F.H. (2011). The DIVA Model: A neural theory of speech acquisition and production. Language and Cognitive Processes, 26(7), 952-981. doi: 10.1080/01690960903498424 Trigos, I., Ysunza, A., Vargas, D., & Vazquez, M. C. (1988). The San Venero Roselli pharyngoplasty: an electromyographic study of the palatopharyngeus muscle. The Cleft Palate Journal, 25(4), 385-388. Troup, G.J., Welch, G., Volo, M., Tronconi, A., Ferrero, F., & Farnetani, E. (1989). On velum opening in singing. Journal of Research in Singing and Applied Vocal Pedagogy, 13(1), 35-39. Ueno, K., Kato, K., & Kawai, K. (2010). Effect of room acoustics on musicians' performance, Part I: Experimental investigation with a conceptual model. Acta Acust. Acust 96, 505–515. Van Demark, D.R., & Hardin, M.A. (1990). Speech therapy for the child with cleft lip and palate. In J. Bardach & H.L. Morris (Eds.), Multidisciplinary management of cleft lip and plate (pp. 799- 806). Philadelphia, PA: W.B. Saunders. Van Lierde, K. M., Claeys, S., De Bodt, M., & Van Cauwenberge, P. (2004). Outcome of laryngeal and velopharyngeal biofeedback treatment in children and young adults: a pilot study. J Voice, 18(1), 97-106. doi: 10.1016/j.jvoice.2002.09.001

144

Van Lierde, K. M., Van Borsel, J., Cardinael, A., Reeckmans, S., & Bonte, K. (2011). The impact of vocal intensity and pitch modulation on nasalance scores: A pilot study. Folia Phoniatrica et Logopaedica, 63, 21-26. doi:10.1159/000319733 Van Lierde, K.M., Wuyts, F. L., Bonte, K., & Van Cauwenberge, P. (2007). The Nasality Severity Index: An objective Measure of Hypernasality Based on a Multiparameter Approach (A Pilot Study). Folia Phonniatrica et Logopaedica, 59, 31-38. doi: 10.1159/000096548 Vennard, W. (1964). An Experiment to Evaluate the Importance of Nasal Resonance in Singing. Folia Phoniatr Logop, 16, 146-153. doi: 10.1159/000262995 Verdolini, K. (2000). Resonant voice therapy. In J.C. Stemple (Ed.), Voice Therapy: clinical studies (pp. 46-62). San Diego, CA: Singular Publishing. Volo, M. Farnetani, E. Troup, G.J. & Ferrero, F.E. (1986). A pilot study of nasal aerodynamic flow in singing. Quaderni del Centro di Studio per le Ricerche di Fonetica del C.N.R. 5, 395-404. Vurma, A., & Ross, J. (2002). Where is a singer's voice if it is placed “Forward”? The Journal of Voice, 16(3), 383-391. doi:10.1016/S0892-1997(02)00109-1 Vurma, A., & Ross, J. (2003). The perception of 'forward' and 'backward placement' of the singing voice. Logopedics Phoniatrics Vocology, 28(1), 19-28. doi:10.1080/14015430310010854 Warren, D.W. (1967). Nasal emission of air and velopharyngeal function. Cleft Palate Journal, 4, 148- 155. Warren, D.W., Duany, L.F., & Fischer, N.D. (1969). Nasal pathway resistance in normal and cleft lip and palate subjects. Cleft Palate Journal, 6, 134–140. Watson, M., & Lof, G.L. (2008). What we know about nonspeech oral motor exercises. Seminars in Speech and Language, 29, 320-330. Watterson, T., & Lewis, K. E. (2006). Test-retest nasalance score variability in hypernasal speakers. Cleft Palate Craniofac J, 43(4), 415-419. doi:10.1597/05-104.1 Watterson, T., Lewis, K. E., & Foley-Homan, N. (1999). Effect of Stimulus Length on Nasalance Scores. The Cleft Palate-Craniofacial Journal, 36(3), 243–247. doi: 10.1597/1545- 1569_1999_036_0243_eoslon_2.3.co_2 Watterson, T., Lewis, K., & Brancamp, T. (2005). Comparison of Nasalance scores obtained with the Nasometer 6200 and the Nasometer II 6400. Cleft Palate Craniofac J, 42(5), 574-579. doi:10.1597/04-017.1

145

Watterson, T., Lewis, K., Allord, M., Sulprizio, S., & O’Neill, P. (2007). Effect of vowel type on reliability of nasality ratings. Journal of Communication Disorders, 40(6), 503–512. doi: 10.1016/j.jcomdis.2007.02.002 Watterson, T., Mancini, M., Brancamp, T.U., & Lewis, K. E. (2013). Relationship between the perception of hypernasality and social judgments in school-aged children. Cleft Palate Craniofac J, 50(4), 498-502. doi: 10.1597/11-126 Watterson, T., York, S.L., & McFarlane, S.C. (1994). Effects of vocal loudness on nasalance measures. Journal of Communication Disorders, 27, 257-262. doi: 10.1016/0021-9924(94)90004-3. Weihs, C., Jannach, D., Vatolkin, I., & Rudolph, G. (Eds.). (2016). Music data analysis: Foundations and applications. Boca Raton, FL: CRC Press. Welch, G., Howard, D. M., & Nix, J. (2019). The Oxford handbook of singing. Oxford, UK; New York, NY: Oxford University Press. White, E.G. (1938). Sinus tone production. London, UK: J.M. Dent & Sons Whitehill, T. L., & Lee, A. S.-Y. (2008). Instrumental analysis of resonance in speech impairment. In M. J. Ball, M. R. Perkins, N. Müller & S. Howard (Eds.), The handbook of clinical linguistics (pp. 332-343). Oxford, UK: Blackwell Publishing. Whitehill, T. L., Gotzke, C. L., & Hodge, M. (2013). Speech intelligibility. In S. Howards & A. Lohmander (Eds.), Cleft palate speech: Assessment and intervention (pp. 293–304). Chichester, UK: Wiley-Blackwell. Whitehill, T. L., Lee, A. S. Y., & Chun, J. C. (2002). Direct magnitude estimation and interval scaling of hypernasality. Journal of Speech, Language, and Hearing Research, 45(1), 80-88. doi: 10.1044/1092-4388(2002/006) Whitener, J.J. (2016). The German School of Singing: A compendium of German treatises 1848-1965 (Unpublished doctoral dissertation). Indiana University, Bloomington, Indiana. Wilbrand, H. (1992). X-ray study of articulation and formant frequencies in two female singers. Journal of Research in Singing and Applied Vocal Pedagogy, 16(1), 29-42. Willadsen, E., Lohmander, A., Persson, C., Lundeborg, I., Alaluusua, S., Aukner, R., Bau, A., Boers, M., Bowden, M., Davies, J., Emborg, B., Havstam, C., Hayden, C., Henningsson, G., Holmefjord, A., Hölttä, E., Kisling-Møller, M., Kjøll, L., Lundberg, M., McAleer, E., Nyberg, J., Paaso, M., Pedersen, N.H., Rasmussen, T., Reisæter, S., Søgaard Andersen, H., Schöps, A., Tørdal, I-B., & Semb, G. (2017). Scandcleft randomised trials of primary surgery for unilateral cleft lip and palate: 5. Speech outcomes in 5-year-olds - consonant proficiency and

146

errors. Journal of Plastic Surgery and Hand Surgery, 51(1), 38-51. doi: 10.1080/2000656X.2016.1254647 Witt, P. D., & D’Antonio, L. L. (1993). Velopharyngeal insufficiency and secondary palatal management. A new look at an old problem. Clinics in Plastic Surgery, 20, 707–721. Witzel, M. A., Tobe, J., & Salyer, K. (1988). The use of nasopharyngoscopy biofeedback therapy in the correction of inconsistent velopharyngeal closure. International Journal of Pediatric Otorhinolaryngology, 15(2), 137-142. doi: 10.1016/0165-5876(88)90064-X Wood, A., & Bowsher, J.M. (1965). The physics of music. London, UK: Methuen. Wooldridge, W.B. (1956). Is there nasal resonance? The NATS Bulletin, 13(1), 28-29. World Health Organization. (2020). International collaborative research on craniofacial anomalies. Retrieved from https://www.who.int/genomics/anomalies/en/ Xu, Y., Larson, C. R., Bauer, J. J., & Hain, T. C. (2004). Compensation for pitch-shifted auditory feedback during the production of Mandarin tone sequences. J Acoust Soc Am, 116(2), 1168- 1178. doi: 10.1121/1.1763952 Yamaoka, M., Matsuya, T., Miyazaki, T., Nishio, J., & Ibuki, K. (1983). Visual training for velopharyngeal closure in cleft palate patients; a fibrescopic procedure (preliminary report). Journal of Maxillofacial Surgery, 11, 191-193. Yamashita, R. P., Borg, E., Granqvist, S., & Lohmander, A. (2018). Reliability of hypernasality rating: Comparison of 3 different methods for perceptual assessment. The Cleft Palate-Craniofacial Journal, 55(8), 1060–1071. doi: 10.1177/1055665618767116 Yanagisawa, E., Estill, J., Mambrino, L., & Talkin, D. (1991). Supraglottic contributions to pitch raising. Videoendoscopic study with spectroanalysis. Ann Otol Rhinol Laryngol, 100(1), 19-30. doi:10.1177/000348949110000104 Yanagisawa, E., Kmucha, S. T., & Estill, J. (1990). Role of the Soft Palate in Laryngeal Functions and Selected Voice Qualities. Annals of Otology, Rhinology & Laryngology, 99(1), 18–28. doi10.1177/000348949009900104 Yorkston K. M., Spencer K., Duffy J., Beukelman D., Golper L. A., Miller R., Strand, E., & Sullivan M. (2001). Evidence-based practice guidelines for dysarthria: Management of velopharyngeal function. Journal of Medical Speech-Language Pathology, 9(4), 257–274. Young, L. H., Zajac, D. J., Mayo, R., & Hooper, C. R. (2001). Effects of vowel height and vocal intensity on anticipatory nasal airflow in individuals with normal speech. J Speech Lang Hear Res, 44(1), 52-60. doi: 10.1044/1092-4388(2001/005)

147

Zajac, D. J. (2000). Pressure-Flow Characteristics of /m/ and /p/ Production in Speakers without Cleft Palate: Developmental Findings. The Cleft Palate-Craniofacial Journal, 37(5), 468–477. doi: 10.1597/1545-1569_2000_037_0468_pfcoma_2.0.co_2 Zajac, D. J., & Vallino, L. D. (2017). Evaluation and management of cleft lip and palate: A developmental perspective. San Diego, CA: Plural Publishing. Zajac, D. J., & Mayo, R. (1996). Aerodynamic and temporal aspects of velopharyngeal function in normal speakers. Journal of Speech and Hearing Research, 39(6), 1199-1207. doi: 10.1044/jshr.3906.1199 Zajac, D. J., Mayo, R., & Kataoka, R. (1998). Nasal coarticulation in normal speakers: A re-examination of the effects of gender. Journal of Speech, Language, and Hearing Research, 41(3), 503-510. doi: 10.1044/jslhr.4103.503 Zalzal, H.G., O’Brien, D.C., & Zalzal, G.H. (2018). Pediatric anatomy: Nose and sinus. Operative Techniques in Otolaryngology-Head and Neck Surgery, 29(2), 44-50. doi:10.1016/j.otot.2018.03.002 Zatorre, R., & Salimpoor, V. (2013). From perception to pleasure: Music and its neural substrates. Proceedings of the National Academy of Sciences of the United States of America, 110, 10430-10437. doi: 10.1073/pnas.1301228110 Zraick, R. I., & Liss, J. M. (2000). A comparison of equal-appearing interval scaling and direct magnitude estimation of nasal voice quality. Journal of Speech, Language, and Hearing Research, 43(4), 979-988.

148

Appendices

Appendix 1.1. & 6.2.

Phonetically-balanced stimulus (Fairbanks, 1960). When the sunlight strikes raindrops in the air, they act like a prism and form a rainbow. The rainbow is a division of white light into many beautiful colours. These take the shape of a long round arch, with its path high above, and its two ends apparently beyond the horizon. There is, according to legend, a boiling pot of gold at one end. People look, but no one ever finds it. When a man looks for something beyond his reach, his friends say he is looking for the pot of gold at the end of the rainbow.

Non-nasal sentence (Fletcher, 1976). Look at this book with us, it’s a story about a zoo.

Nasal-loaded sentence (Fletcher, 1976). Mama made some lemon jam.

Appendix 6.1. Instructional protocol for influence of voice focus adjustments on oral-nasal balance in speech and song.

Materials - Water, straight-back chair, copies of all reading passages, copy of musical passage, musical accompaniment audio file, music stand, two pairs of headphones, headphone jack splitter, audio recorder, nasometer Posture - Sit up straight with your feet flat on the floor Warm-up - Read passages aloud - Sing along several times with accompaniment track to orient yourself to the tune Effort - Mezzo-forte (medium loud)

Protocol

Backward voice focus: PRIMING 1. Tell the participant “we will place the voice downward and backward creating a large hollow space for our voice to resonate in” PROCEDURE MODELLING AND INSTRUCTION 2. Open the mouth wide and inhale on the gesture of a yawn-sigh to maximize laryngeal lowering (use a finger to monitor its position). 3. Maintain this lowered laryngeal position while slowly breathing in and out several times (resist letting the larynx reflexively relax) 4. With a retracted tongue, say /a/ on a descending glissando attempting to maintain the lowered laryngeal position/exaggerated hollow throat (as if speaking through a stifled yawn, in a dark, throaty operatic voice)

149

5. Continue to pay attention to producing /a/ in this manner. Try not to relax the posture. Use the yawn to reset the vocal tract as many times as necessary. PROCEDURE REHEARSAL 6. Ask the participant to join in and imitate you. If the participant falters, ask them to stop, and produce the model for them again. Eventually fade out the clinician’s voice and only silently mirror the model with the participant. Depending on how the participant is doing, the clinician may have to fade their voice back in from time to time, during this learning stage of the protocol. RESEARCH TASK 7. Ask the participant to speak/sing the stimuli provided attempting to maintain the lowered laryngeal position/exaggerated hollow throat (as if speaking through a stifled yawn or in a dark, throaty operatic voice). 8. Continue to encourage the participant to pay attention to speaking/singing in in this manner, and to try not to relax the posture. The participant may use the yawn to reset the vocal tract as many times as necessary.

Forward voice focus: PRIMING 1. Tell the participant “we will place the voice forward and up creating a small constricted space for our voice to resonate above our larynx” PROCEDURE MODELLING AND INSTRUCTION 2. Swallow to maximize laryngeal elevation (use a finger to monitor its position). 3. Maintain this elevated laryngeal position while slowly breathing in and out several times (resist letting the larynx reflexively relax) 4. With a forward thrust tongue, say /i/ on an ascending and descending glissando attempting to maintain the raised laryngeal position/exaggeratedly constricted throat (as if saying “cheese” when smiling for a picture, in a bright child-like cartoon voice). 5. Continue to pay attention to producing /i/ in this manner. Try not to relax the posture. Use the swallow to reset the vocal tract as many times as necessary. PROCEDURE REHEARSAL 6. Ask the participant to join in and imitate you. If the participant falters, ask them to stop, and produce the model for them again. Eventually fade out the clinician’s voice and only silently mirror the model with the participant. Depending on how the participant is doing, the clinician may have to fade their voice back in from time to time, during this learning stage of the protocol. RESEARCH TASK 7. Ask the participant to speak/sing the stimuli provided attempting to maintain the raised laryngeal position/exaggeratedly constricted throat (as if saying “cheese” when smiling for a picture in a bright child-like cartoon voice). 8. Continue to encourage the participant to pay attention to speaking/singing in in this manner, and to try not to relax the posture. The participant may use the swallow to reset their vocal tract as many times as necessary.

150

A portion of the contents of Chapter 2 “Control of oral-nasal balance in speech and song” was published in The Oxford Handbook of Music and the Brain (2019). Permission to reproduce the text in this thesis was granted by Oxford Publishing Limited (License date: July 17, 2019; ISSN # 9780198804123; License #16382). Chapter 3 “Influence of altered auditory feedback on the regulation of oral-nasal balance in song” was published in The Journal of Voice (2018). Permission to reproduce the text in this thesis was not required by Elsevier Canada. Chapter 6 “Influence of voice focus adjustments on the regulation of oral-nasal balance in speech and song” was published in Folia Phoniatrica et Logopaedia (2019). Permission to reproduce the text in this thesis was granted by S. Karger AG (License date: September 9, 2019; ISSN #1421- 9972; License #4664950311663). Chapter 7 “Influence of voice focus adjustments on hypernasal speakers’ nasalance scores” was published in The International Journal of Pediatric Otorhinolaryngology (2020). Permission to reproduce the text in this thesis was not required by Elsevier Canada. The PhD candidate was the primary author and contributor to all publications, as described.

151