The Low Mandible Maneuver: Preliminary Study of Its Effects on Aerodynamic and Acoustic Measures

ARTICLE IN PRESS

The Low Mandible Maneuver: Preliminary Study of Its Effects on Aerodynamic and Acoustic Measures Eve Mercer, and Soren Y. Lowell, Syracuse, New York Summary: Objectives. The purpose of this preliminary study was to determine the aerodynamic and acoustic effects of the low mandible maneuver (LMM) as compared to normal voice production. Methods. Ten participants with normal voice characteristics who were nonsingers produced sustained vowel and repeated syllable utterances during two different speaking conditions: using the LMM and using normal phonation posture. The LMM involves a wider vocal tract configuration with a lowered and relaxed jaw position. Acoustic recordings and analyses were performed to determine formants 1 and 2 (F1 and F2) and sound pressure level. Aerodynamic data were collected and analyzed to investigate the effects of the LMM on mean peak pressure, mean airflow, aerodynamic power, aerodynamic efficiency, and aerodynamic resistance. Results. Participants showed greater aerodynamic efficiency, mean peak pressure, and sound pressure level during the LMM condition as compared to normal phonation. The LMM vocal tract configuration changes were also associated with a lowering of F1 and F2 relative to normal voice production. Conclusions. The lowering of the mandible and increased oral area that occurred during the LMM increased vocal efficiency and sound output without significant change to parameters that can be associated with increased vocal effort. These changes in filter configuration were associated with changes in vocal tract resonances. The LMM was readily learned and implemented by healthy participants in this study, and may have utility for singers in training as well as people with hyperfunctional voice disorders. Key Words: Low mandible maneuver−Singing−Voice resonance−Vocal tract−Aerodynamic−Acoustic.

INTRODUCTION The low mandible maneuver (LMM) is a technique that was described in professional singers as a means of enhancing vocal output by altering laryngeal, pharyngeal, and oral configurations.1 The physiologic components of the LMM, including greater mouth opening, prominent mandibular lowering, and lowering of the larynx, are characteristics that were observed in top-ranked professional singers as strategies for enhancing vocal resonance. Through engaging the temporomandibular joint to lower the posterior and anterior portions of the mandible, the tongue is lowered and the overall space in the oral cavity and surrounding regions is markedly increased. These area changes are accompanied by objectively measured lowering of laryngeal position.1 The LMM is instructed or trained by having participants produce the initial posture of a yawn. Therefore, the targeted physiology of the LMM includes that of the yawnsigh technique used in the treatment of hyperfunctional voice disorders. Laryngeal lowering is accompanied by pharyngeal widening in the yawn-sigh technique,2 and prior studies suggest that both the LMM and the yawn-sigh techniques involve widening and lengthening of multiple areas of the vocal tract from the larynx to the lips. The filtering effects of the vocal tract on the sound source are well established,3 and manipulations to the configuration of any Accepted for publication December 7, 2018. From the Department of Communication Sciences and Disorders, Syracuse University, Syracuse, New York. Address correspondence and reprint requests to Soren Y. Lowell, Department of Communication Sciences and Disorders, Syracuse University, 621 Skytop Rd, Suite 1200, Syracuse, NY 13244. E-mail: [email protected] Journal of Voice, Vol. &&, No. &&, pp. &&−&& 0892-1997 © 2018 The Voice Foundation. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jvoice.2018.12.005

region along the length of the vocal tract will impact resulting resonances. Coupling of narrow regions such as the adducted vocal folds to wider regions in the vocal tract can enhance vocal resonance such that the energy of certain harmonics is amplified,1 and overall vocal economy and vocal intensity are optimized.4 Therefore, the multiple changes to vocal tract configuration associated with the LMM may positively impact vocal resonance by enhancing filter function or source−filter interactions during phonation. Physiologic strategies for achieving pitch and resonance targets such as the singers’ formant have been studied in professional singers.1,5 Nair et al used magnetic resonance imaging and ultrasound to quantitatively assess the position changes that occurred from rest to singing posture with the LMM in vowels produced by five professional singers, and to determine the associated acoustic effects of these vocal tract changes. Substantial lowering of the jaw and the position of the larynx were observed during the sung LMM condition as compared to speaking, with acoustic effects that included a pronounced increase in sound pressure level (SPL) and increased energy of multiple harmonics and the singer's formant.1 Prior studies suggest that laryngeal lowering may help professional singers to achieve an elongated vocal tract with a wider pharyngeal cavity, a sound filter configuration that may be essential to the singing formant.6,7 Furthermore, articulatory targets of jaw lowering and altered lingual configuration are strategies implemented by singers for formant tuning and increased overall SPL.5,8 Laryngeal lowering, pharyngeal widening, and oral cavity widening are physiologic components of several treatment approaches that have been successfully applied to people with voice disorders. These combined changes can lengthen the vocal tract and create impedance matching during

ARTICLE IN PRESS 2 vocalization, variables that are associated with decreased vocal fold adduction9,10 and loosely approximated vocal folds with reduced collision forces.11 These targets are therefore important in the treatment of hyperfunctional dysphonia, which often involves habitually increased adductory and collision forces during phonation.12,13 Circumlaryngeal massage treatment includes goals of laryngeal lowering and reduction of tension in the suprahyoid muscles, and can result in improved auditory-perceptual and acoustic measures of voice.14−16 The yawn-sigh treatment technique targets a lowered laryngeal position and widening of the oropharyngeal vocal tract region in the yawn posture, followed by a breathy phonation onset. The physiologic targets of the yawn-sigh are considered appropriate for treating hyperfunctional dysphonia by decreasing patterns of overactivation of the intrinsic and extrinsic laryngeal muscles while optimizing vocal output.2 Videoendoscopic recordings and magnetic resonance imaging measurements in healthy subjects show that the yawn-sigh or yawny voice quality is associated with pharyngeal widening, laryngeal lowering, increased vocal tract area, and vocal tract lengthening.2,17,18 Because of the alterations to vocal tract configuration that occur with a yawn maneuver, the acoustic effect of the yawn-sigh technique and yawny quality on formants 1−3 have been studied as a means of measuring source−filter changes.2,17 Variable results have been reported, with a lowering of F1 and F2 during an /ɑ/ vowel task produced with yawny voice quality evidenced in one study,17 whereas a lowering of F2 and F3 with minimal changes in F1 were findings reported in another study that included the yawn configuration plus a sigh (breathy release) and /i/ vowel vocalization.2 Changes to formants and formant transitions have been documented following treatment that targets a lowered laryngeal position in people with hyperfunctional voice disorders.19−21 Determination of whether formant frequencies are affected during the LMM as compared with typical phonation would provide an objective indicator of the acoustic filtering effects associated with the LMM. Semioccluded vocal tract configuration is a resonancefocused therapeutic approach that is also frequently applied to people with hyperfunctional voice disorders, and often includes a target of pharyngeal widening that can be associated with improved voice quality.22,23 One type of semioccluded vocal tract configuration treatment, resonant voice therapy, often implements speech stimuli with nasal, bilabial consonants and low vowels such as “molm” to achieve a wider oral cavity area and a rounded, more occluded lip configuration to facilitate forward resonance focus and sensory feedback.24 These changes in vocal tract configuration are thought to optimize the source−filter interaction by altering the interaction between harmonics and associated formants,11,25 with a resulting voice that provides functional loudness without involving increased vocal fold adductory forces to achieve that loudness.9,26 Loudness level is frequently lowered in people with dysphonia,27 and not being well heard or understood by others substantially contributes to the negative impact of a voice disorder.28 Therefore, therapeutic approaches that optimize loudness level while

Journal of Voice, Vol. &&, No. &&, 2018

potentially minimizing adductory forces during voicing are important in the treatment of dysphonia. Aerodynamic changes are associated with vocal tract configuration differences and relate to the physiologic and acoustic changes that have been the focus of study in prior research addressing the LMM. The lowering of the entire mandible that is instructed in the LMM may contribute to the documented physiology of this maneuver, including a more open oral and pharyngeal cavity with rounded (semioccluded) lip configuration and a lowered laryngeal position.1 Several of these vocal tract changes are known to increase the inertance in the vocal tract, which has been associated with aerodynamic and acoustic changes that include increased oral pressure, decreased phonation threshold pressure, and greater intensity of higher harmonics in output.4,25,29 Titze describes the beneficial effects of semioccluded vocalization techniques on the basis of their ability to strengthen the interactions between the source and the filter, thereby increasing vocal intensity, efficiency, and economy.4 Based on aerodynamic findings related to semioccluded techniques, the similar physiologic features of the LMM should produce aerodynamic and acoustic changes characterized by increased vocal efficiency and output. A highly efficient voicing style would optimize vocal output while minimizing energy expenditure. While the LMM has been characterized as a strategy used by professional singers to optimize singing resonances, its physiologic targets could be beneficial for people with voice disorders. By achieving the oropharyngeal and laryngeal vocal tract configuration that is associated with the yawnsigh technique while focusing on a pronounced lowering of the entire mandible, the LMM could optimize vocal resonance without necessitating the breathy “sigh” component that can reduce loudness and can be initially difficult to achieve for people with pronounced adductory hyperfunction. Aerodynamic changes associated with the LMM and how those may relate to acoustic changes are not yet understood. Determining the aerodynamic features of the LMM would be important to effectively train the LMM to novice singers or people with voice disorders for improved vocal output. The purpose of this preliminary study was to determine the aerodynamic and acoustic effects of the LMM in nonsingers. In this initial investigation of aerodynamic physiology and acoustics associated with the LMM, it was important to study people with normal voice characteristics that were not already altered by functional or structural voice disorders. METHODS Participants This study was approved by the Institutional Review Board at Syracuse University. All participants were paid for their participation and provided informed consent prior to beginning the study. Thirteen females ages 21−38 years (mean age 23.9) were recruited from the Syracuse University campus through flyers and email postings. All participants met the following criteria as determined through a phone

ARTICLE IN PRESS Eve Mercer and Soren Y. Lowell

The Low Mandible Maneuver: Preliminary Study of Its Effects

screening: (a) over the age of 18 years; (b) in good general health per self-report; (c) used English as their primary language; (d) had no known hearing loss; (e) were nonsmokers for at least the last 5 years; (f) had no voice or swallowing problems per self-report; and (g) demonstrated appropriate overall voice quality per informal, binary judgment of connected speech during the telephone screening and on the day of testing. Additionally, none of the participants were professional singers. Female participants only were included in this initial study to avoid gender differences in aerodynamic measures that may skew results with a small overall sample size. Data from 10 of the 13 participants were analyzed, with the remaining datasets excluded for the following reasons: voice quality was judged as normal during the phone screening but abnormal on the day of testing, professional voice experience was determined subsequent to the phone screening, and one participant was unable to appropriately perform the LMM despite cues and training.

Procedures Training and practice were provided on two different speaking conditions: normal phonation and phonation with the LMM. For the LMM, participants were instructed to act as though they were beginning to yawn, but to stop when their mouth was maximally open in the yawn sequence. Then, using the LMM position, they were instructed to produce a sustained /ɑ/ or repeated set of /pɑ/ syllables for acoustic and aerodynamic recordings. Participants were instructed to focus on two targets: a lowering and relaxing of the entire jaw, and a wider mouth opening than one would perform in conversational speech. Visual and auditory modeling by the researchers was incorporated into each training session. For each participant, at least three practice trials were elicited for both the /ɑ/ and /pɑ/ productions, with feedback provided on the accuracy of use of the LMM physiologic targets. The Computerized Speech Lab (KayPENTAX, Montvale, New Jersey) was used to collect acoustic data. A head-mounted condenser microphone was placed at a 9-cm

3

distance from the corner of the mouth, and at a 45° angle. Three normal phonation sustained /ɑ/ utterances, approximately 5−6 seconds each, were elicited at a comfortable pitch and loudness level with breaths in between each utterance. Then, the same procedure was followed but with the use of the LMM for obtaining the three tokens of the sustained /ɑ/. The participant's initial /ɑ/ recording in the normal condition was replayed when the LMM was elicited, so that the participant could match their initial, comfortable pitch and loudness to maintain consistency across all conditions. To quantitatively verify differences in mouth opening across conditions, digital video recordings were obtained for each participant during these vocalizations. One still image from each of the three /ɑ/ tokens for both the normal phonation and LMM conditions was selected and analyzed using a pixel-based software program (ImageJ, National Institutes of Health, Bethesda, Maryland). Area measurements were computed and a mean of the three values for each participant in the two conditions was generated. See Figure 1 for example images for one participant. Mean area measurements for degree of mouth opening across all participants were more than twice as large for the LMM condition (4,229.5 pixels for normal phonation versus 10,073.6 pixels for LMM phonation), verifying that participants achieved the targeted physiology. The aerodynamic data were obtained during repeated /pɑ/ productions using the Phonatory Aerodynamic System (PAS, KayPENTAX, Montvale, New Jersey) and following established procedures.30,31 A facial mask covering the nose and mouth was placed and held securely by the participant. Air pressure and airflow levels were measured through the PAS system, and a calibrated microphone built into the pneumotachometer system allowed for SPL recordings measured directly in front of the airflow output. The participants were provided a model and instructed to produce three sets of successive, connected /pɑ/ syllable repetitions (approximately seven syllables), with a breath between each set. As with the acoustic data collection procedures, tokens were elicited first in the normal phonation condition and then in the

FIGURE 1. Still images from digital video recordings of the sustained /ɑ/ task under the normal phonation and low mandible maneuver (LMM) conditions.

ARTICLE IN PRESS 4

Journal of Voice, Vol. &&, No. &&, 2018

LMM phonation condition, following retraining on the maneuver. The initial recording of the normal phonation condition of the sustained /ɑ/ was replayed for each participant to remind them of their initial, comfortable pitch and loudness level. Three or more tokens of the /pɑ/ sequences were elicited and monitored throughout data collection to ensure viable pressure peaks and continuous phonation.

Acoustic and aerodynamic analyses From the acoustic recordings, data were analyzed to determine the first and second formant frequencies and SPL. F1 and F2 are the formants that are most affected by oral configuration changes,3 and are considered important variables for reflecting the effects of vocal tract configuration changes in yawn-related vocalization2,17 and in manual circumlaryngeal treatment.20,21 However, the effect of the LMM on these formants is unknown. Voice intensity was important to determine as it is known to be affected by resonance changes and source−filter interactions, and increased SPL that involves low vocal fold stress or phonatory effort is a target of several therapeutic techniques.4,10 Sustained /ɑ/ vowels were used to extract formant measures. For each /ɑ/ production, analysis began 2.0 seconds into the recording to avoid any onset effects. Mean F1 and F2 values were determined for each participant across the three /ɑ/ tokens, and mean SPL values were generated for each participant based on the analyzed syllable sequences. F1 and F2 were analyzed using Praat software32 with an analysis window of 0.150 seconds, while SPL was obtained from the aerodynamic system recordings. Measures of peak pressure, aerodynamic efficiency, aerodynamic power, and laryngeal resistance were of interest in this study because prior research suggests that they are therapeutically important when training techniques that address altered vocal tract configuration.4 Dependent measures were derived from the aerodynamic data using the Voicing Efficiency software within the PAS (KayPENTAX, Montvale, New Jersey), and included mean peak air pressure, mean airflow during voicing, aerodynamic power (peak pressure*target airflow*standard factor), aerodynamic resistance, and aerodynamic efficiency (SPL-related acoustic power/aerodynamic power). Vocal efficiency represents the effective SPL that is generated from aerodynamic energy, and may be an important indicator

of vocal economy in its reflection of output power relative to aerodynamic power. Syllable peak sequences were selected on the basis of similarity of peak measurements. Sequential peaks that were within 0.5 cm of each other were preferentially selected, but for some a margin of 1 cm was allowed in order to collect sufficient data. For most participants, at least three /pɑ/ sequences were analyzed to derive the above-noted dependent measures, and each analyzed portion excluded the first syllable and included at least three syllables. Statistical analyses IBM SPSSv24 (Armonk, New York) was used for statistical analyses for both the acoustic and aerodynamic data. Preliminary testing of the data distributions showed that one aerodynamic measure (aerodynamic efficiency) and two acoustic measures (SPL and F1) were not normally distributed. Therefore, the nonparametric-related samples Wilcoxon signedrank test was used to determine statistically significant differences between conditions for those measures (paired comparison testing), whereas differences between conditions for all other measures were tested with paired-samples t tests. Significance level was defined at P < 0.05. RESULTS Overall mean acoustic measurements for the normal phonation and LMM conditions are presented in Table 1, with mean aerodynamic measurements displayed in Table 2. Dependent measures marked with an asterisk (*) showed statistically significant differences between the normal phonation condition and the LMM condition. Acoustic measures Comparisons between the two conditions showed that mean values for F1 (P = 0.009, Cohen's d = 1.33) and F2 (P ≤ 0.001, Cohen's d = 2.08) were significantly lower during the LMM condition as compared to the normal phonation condition (Table 1). Most participants had lower values for F1 during the LMM condition as compared to normal phonation (Figure 2), and all participants had lower values for F2 during the LMM condition (Figure 3). Additionally, SPL

TABLE 1. Means (Standard Deviations), Confidence Intervals, and Significance Levels for Acoustic Measures in the Normal Phonation Condition and LMM Condition. Normal Phonation Condition Mean (SD) Mean SPL voicing (dB) 81.22 (2.99) Formant 1 (Hz) 792.44 (114.99) Formant 2 (Hz) 1,432.43 (132.33)

LMM Phonation Condition

Confidence Interval (95%)

Mean (SD)

79.08−83.36 710.17−874.70 1,337.77−1,527.10

84.66 (3.38) 709.24 (109.56) 1,206.68 (119.30)

* Denotes statistically significant values at P < 0.05. Abbreviations: LMM, low mandible maneuver; SD, standard deviation.

Confidence Interval (95%) P Value 82.24−87.08 630.86−787.62 1,121.33−1,292.02

0.013* 0.009* <0.001*

ARTICLE IN PRESS Eve Mercer and Soren Y. Lowell

5

The Low Mandible Maneuver: Preliminary Study of Its Effects

TABLE 2. Means (Standard Deviations), Confidence Intervals, and Significance Levels for Aerodynamic Measures in the Normal Phonation Condition and LMM Condition. Normal Phonation Condition

Mean peak pressure (cm H2O) Mean airflow voicing (L/s) Aerodynamic power (watts) Aerodynamic resistance (cm H2O [L/s]) Aerodynamic efficiency (p.p.m.)

LMM Phonation Condition

Mean (SD)

Confidence Interval (95%)

Mean (SD)

Confidence Interval (95%)

P Value

7.76 (1.39) 0.17 (0.04) 0.14 (0.05) 45.65 (6.84) 165.95 (85.38)

6.77−8.76 0.14−0.20 0.10−0.17 40.76−50.54 104.88−227.03

8.42 (1.49) 0.17 (0.04) 0.15 (0.06) 49.80 (10.52) 361.08 (202.26)

7.36−9.49 0.14−0.20 0.11−0.19 42.28−57.32 216.39−505.76

0.007* 0.809 0.109 0.117 0.017*

* Denotes statistically significant values at P < 0.05. Abbreviations: LMM, low mandible maneuver; SD, standard deviation.

FIGURE 2. Formant 1 (F1) for individual participants, measured during the sustained /ɑ/ task under the normal phonation and low mandible maneuver (LMM) conditions. Mean values for the three /ɑ/ tokens are shown for each participant.

FIGURE 3. Formant 2 (F2) for individual participants, measured during the sustained /ɑ/ task under the normal phonation and low mandible maneuver (LMM) conditions. Mean values for the three /ɑ/ tokens are shown for each participant.

ARTICLE IN PRESS 6

Journal of Voice, Vol. &&, No. &&, 2018

FIGURE 4. Mean peak pressure for individual participants, measured during the repeated /pɑ/ syllable task under the normal phonation and low mandible maneuver (LMM) conditions. Mean values for the analyzed tokens are shown for each participant. values were significantly higher in the LMM condition as compared to the normal phonation condition (P = 0.013, Cohen's d = 1.14), with mean values that were approximately 3.5 dB higher during the LMM (Table 1).

Aerodynamic measures Comparisons between the two conditions showed that mean values for mean peak pressure (P = 0.007, Cohen's d = 1.09) and aerodynamic efficiency (P = 0.017, Cohen's d = 0.95) were significantly greater during the LMM condition as compared to the normal phonation condition (Table 2). The majority of participants showed higher peak pressures during the LMM condition (Figure 4), and virtually all of the participants showed higher aerodynamic efficiency during the LMM condition as compared to normal phonation

(Figure 5). In contrast, the aerodynamic measures of mean airflow during voicing (P = 0.809), aerodynamic power (P = 0.109), and aerodynamic resistance (P = 0.117) did not show significant differences between the normal phonation and LMM conditions (Table 2). DISCUSSION Whereas a prior study had investigated the singing-related configuration changes and associated acoustics that characterize the LMM in professional singers,1 the current study examined the effects of the LMM during speaking tasks in nonsingers. This study served as a preliminary step toward determining the potential utility of the LMM as a vocal resonance-enhancing maneuver, which may be advantageous as a training technique for novice singers or for treating

FIGURE 5. Aerodynamic efficiency for individual participants, measured during the repeated /pɑ/ syllable task under the normal phonation and low mandible maneuver (LMM) conditions. Mean values for the analyzed tokens are shown for each participant.

ARTICLE IN PRESS Eve Mercer and Soren Y. Lowell

The Low Mandible Maneuver: Preliminary Study of Its Effects

hyperfunctional voice disorders. Several acoustic and aerodynamic voice parameters showed significant change when participants vocalized while using the LMM as compared with a normal speaking posture. These findings advance our understanding of the effects of the LMM beyond the realm of singers and suggest that its benefits are applicable to connected speech.

Acoustic effects of the LMM In the present study, formants 1 and 2 were both significantly lower in the LMM condition as compared to the normal phonation condition. Alterations in filter function with the LMM are presumed due to the associated vocal tract configurations, but have not been previously objectively quantified with F1 and F2 variables. In a prior study that included acoustic analysis of the /ɑ/ and /i/ vowels during the LMM, graphic examples of harmonic amplitudes were provided for single utterances of four subjects.1 However, comparative data to show differences between typical-phonation /ɑ/ and the LMM /ɑ/ were not reported, and mean data to show overall acoustic results for the group were also not presented. The lowering of F1 and F2 that occurred in the present study is consistent with findings by Story et al, who found marked lowering of F1 and F2 on a spoken /ɑ/ vowel when a yawny voice quality was used by nonsingers. The yawn-sigh has been associated with a lowering of F2 with no change in F1 in nonsingers, but these changes were reported for an /i/ vowel that is likely to differ from changes that occur with an /ɑ/ vowel.2 In the present study, the LMM created a lower jaw and tongue position with increased lip rounding. Although the source−filter theory describes a general increasing effect on F1 as tongue height is lowered, these effects are lessened on back vowels such as the /ɑ/ used in this study.3 Furthermore, pharyngeal configuration and lip rounding both affect F1; widening of the pharynx is expected to decrease F133 and greater lip rounding lowers both F1 and F2.3 Prior physiologic measurements of laryngeal and pharyngeal position during the LMM or a yawn configuration indicate substantial pharyngeal widening,2,34 and our visual, quantitative analysis of lip configuration in the current study showed marked lip rounding and opening associated with the LMM. Thus, the pharyngeal widening and lip rounding that occurred during the LMM condition may contribute to the lowering of F1 and F2. The significant frequency shifts that occurred for F1 and F2 in this study confirm the effect of the LMM on sound filtering and oral resonances during speech. Several prior studies support the relationship between formant change and voice quality improvement in people with voice disorders. Roy et al20 determined pre- and post-treatment mean F1 and F2 values for 111 females with muscle tension dysphonia, measured during vowels that included /ɑ/ . Formant change was also depicted through vowel space area measurements at both time points. Treatment included manual laryngeal reposturing and circumlaryngeal massage,

7

and ratings of auditory-perceptual voice quality were also determined for pre- and post-treatment time points. A significant lowering of F2 occurred after treatment, and an overall lowering of F1 and F2 was observed in several vowels posttreatment, reflected as an expansion and shifting of the vowel space quadrilateral. These formant changes were accompanied by significant improvement in perceptually rated voice quality. It was suggested that changes to hyolaryngeal position, base of tongue configuration, and pharyngeal constriction may have contributed to the lowering of F2 evidenced on /ɑ/ and /u/.20 In an earlier study, Roy and Ferguson showed significant pre- to post-treatment lowering of F1, F2, and F3 measured during a sustained /ɑ/ in 75 participants with muscle tension dysphonia who underwent manual circumlaryngeal treatment. Perceptual ratings of dysphonia severity also showed significant pre- to post-treatment improvement. Roy et al concluded that laryngeal lowering and subsequent vocal tract lengthening were likely contributors to the formant changes, and were consistent with predictions from the source−filter theory.21 Additionally, the authors suggested that biomechanical linkages between laryngeal, lingual, and mandibular muscles and structures may facilitate widespread changes in vocal tract configuration after circumlaryngeal massage techniques, thus contributing to resonance changes in voice output.20,21 In a study of healthy participants without voice problems, difference values for F1−F0 were compared for comfortable (typical) /ɑ/ phonation versus phonation with seven variants of semioccluded vocal tract configuration exercises,10 which like the LMM are techniques that are associated with vocal tract lengthening, pharyngeal widening, and semiocclusion of the lips.22,23 The authors found that all semioccluded techniques resulted in substantially lower F1 values, as evidenced by reduced F1−F0 differences (without significant increase in F0).10 Thus, the resonant voice characteristics targeted in a number of semioccluded vocal configuration treatment techniques are associated with a concomitant lowering of F1. Considering the results of collective studies investigating the effects of circumlaryngeal massage and semioccluded vocal tract exercises, it appears that a lowering of F1 or F2 is associated with improved voice quality and vocal economy. Determining the acoustic changes that are associated with the LMM in people with voice disorders, and how those may relate to changes in voice quality following treatment, is an important area for future investigation. The higher SPL mean value observed with the LMM condition verifies that output sound energy was enhanced with this maneuver. Our SPL findings are consistent with the findings of Nair et al who showed much higher SPL values for sung vowels, which included the LMM than regular spoken vowels without the use of the LMM.1 However, singing is often produced at a naturally higher SPL, so it is difficult to separate the effects of the LMM on SPL from the overall singing effects in that study. In the present study, because both conditions were produced in a speaking versus singing context, the LMM rather than singing effects appear to

ARTICLE IN PRESS 8 account for the SPL differences. Based on the resonance changes that occurred for F1 and F2, as well as the measured changes in oral cavity configuration, the boost in SPL during the LMM suggests that this maneuver provided enhanced filtering or source−filter interactions for spoken voice. Vocal economy has been described as a desirable target that involves high SPL output with low adductory stress at the level of the vocal folds.10,35 Altered upper vocal tract configurations achieved with semioccluded vocal tract techniques can produce more efficient vibratory patterns that result in optimal conversion of subglottic pressure to vibratory amplitude, resulting in increased output intensity.35 Thus, it is possible that the oral configuration changes of the LMM created source−filter interactions that were associated with the increased SPL. Difficulty projecting the voice and achieving functional loudness level is common with hyperfunctional voice disorders,27,36 and reduced loudness level in everyday communication has a significant impact on quality of life in many different voice disorders including those that involve hyperfunction.28 Techniques that may enhance functional loudness level without requiring laryngeal adductory effort can therefore have high treatment utility.

Aerodynamic effects of the LMM Aerodynamic efficiency was of interest in this study because it represents the effective SPL that is generated from aerodynamic energy, and represents the amount of resulting sound energy after normalizing for air pressure and airflow as computed in the Voicing Efficiency software of the PAS. Measures reflecting vocal efficiency or vocal economy are considered important for demonstrating the potential benefit of resonance-altering treatments to professional voice users and people with voice disorders.4,10 Prior studies addressing the physiology of yawn-style vocalization and the LMM have objectively documented vocal tract configurations that are longer and wider in several regions relative to normal vocalization.1,2,17,34 This sets up an impedance-matching source −filter interaction that may positively affect vibratory behavior and vocal output.4,29 In the present study, there was a significant increase in aerodynamic efficiency when using the LMM as compared to a normal speaking posture. Therefore, acoustic energy was increased to a greater degree than aerodynamic energy when participants achieved the lowered jaw position and larger oral cavity configuration that was associated with the LMM. These findings suggest that the vocal tract configuration changes that occur with the LMM may help to optimize acoustic output energy. An important implication of these findings is that the LMM could be used to maximize vocal efficiency by boosting acoustic output without increasing vocal effort. In addition to being critical for singers who must maintain high acoustic output levels for long periods of time, these vocal efficiency effects of the LMM may be advantageous in addressing voice therapy goals for people with hyperfunctional voice disorders. In addition to aerodynamic efficiency, mean peak pressure was significantly higher in the LMM condition than in

Journal of Voice, Vol. &&, No. &&, 2018

the normal phonation condition. Mean peak pressure is measured at the lips with subsequent interpolation of translaryngeal pressure,30,31 and therefore reflects both translaryngeal pressure at the level of the vocal folds as well as oral cavity pressure. The significant increase in mean peak pressure that occurred during the LMM procedure may be associated with either translaryngeal pressure changes, oral pressure changes, or both. Because the measure of laryngeal resistance did not show a significant increase during the LMM condition, it appears that laryngeal valving changes did not primarily account for the increase in this measure. Given the concomitant, significant increase in aerodynamic efficiency and the measured changes in oral cavity configuration, it is reasonable to suspect that changes in oral resonances achieved with the LMM contributed to the increased oral pressure. Mean airflow and aerodynamic resistance both did not show significant differences between the LMM and the normal phonation conditions. An increase in airflow during voicing may involve increased respiratory effort or an overall increase in initial lung volume, neither of which were targeted by the LMM. Therefore, the lack of a significant difference in this measure provides support for the concept that the primary effect of the LMM involved filter effects or source−filter interactions, without predominant changes in respiratory effort. Similarly, the nonsignificant differences in aerodynamic resistance when comparing the LMM to regular phonation suggest that increases in vocal fold adduction patterns (which will result in an increase in laryngeal resistance) did not primarily account for the increase in SPL that occurred during the LMM condition. Thus, the physiology associated with the LMM may be optimal for both singers and people with hyperfunctional voice problems; the lowering of the mandible and open-mouth position can be achieved with minimal effort expenditure and without a concomitant increase in respiratory or laryngeal effort, while the efficiency of vocal output is maximized. CONCLUSIONS In this preliminary study, the LMM was associated with increased aerodynamic efficiency, increased peak pressure, increased SPL, and the lowering of F1 and F2 in participants with normal voice who were nonsingers. Our results suggest that when the LMM is used, the efficiency of vocal output is enhanced through an alteration of vocal tract resonances, which may be beneficial not only to singers with high vocal needs but also to people with voice disorders. Taken together, the demonstrated physiology of the LMM and its initial yawn-like posture may address several physiologic targets in hyperfunctional voice disorders while optimizing vocal power and minimizing vocal effort. Further research is needed to simultaneously assess laryngeal, pharyngeal, and oral configuration changes that may contribute to the LMM, so that the maneuver could be effectively trained as a singing technique or therapeutic technique. This preliminary study incorporated a

ARTICLE IN PRESS Eve Mercer and Soren Y. Lowell

The Low Mandible Maneuver: Preliminary Study of Its Effects

small sample size of 10 participants; a larger study group would help to account for individual variability in anatomy and physiology that may contribute to results, and would promote the generalizability of findings. An important next investigative step is the direct measurement of changes in one or more physiologic parameters associated with the LMM in a group of participants with hyperfunctional voice disorders. Specifically, a comparison of the effects of the LMM to the yawn-sigh technique in people with hyperfunctional dysphonia would yield findings with high clinical utility. The findings from the current study support the rapid trainability of the LMM technique and its immediate aerodynamic and acoustic effects on the speaking voice in people with normal voice characteristics. The enhancement of vocal efficiency during the LMM without apparent increases in vocal effort suggests that the LMM may be therapeutically useful. REFERENCES 1. Nair A, Nair G, Reishofer G. The low mandible maneuver and its resonential implications for elite singers. J Voice. 2016;30:128.e113−132. 2. Boone DR, McFarlane SC. A critical view of the yawn-sigh as a voice therapy technique. J Voice. 1993;7:75–80. 3. Stevens K, House A. Development of a quantitative description of vowel articulation. J Acoust Soc Am. 1955;27:484–493. 4. Titze IR. Voice training and therapy with a semi-occluded vocal tract: rationale and scientific underpinnings. J Speech Lang Hear Res. 2006;49:448–459. 5. Sundberg J, Skoog J. Dependence of jaw opening on pitch and vowel in singers. J Voice. 1997;11:301–306. 6. Sundberg J. Articulatory interpretation of the “singing formant”. J Acoust Soc Am. 1974;55:838–844. 7. Shipp T, Izdebski K. Letter: vocal frequency and vertical larynx positioning by singers and nonsingers. J Acoust Soc Am. 1975;58:1104–1106. 8. Austin SF. Jaw opening in novice and experienced classically trained singers. J Voice. 2007;21:72–79. 9. Verdolini K, Druker DG, Palmer PM, et al. Laryngeal adduction in resonant voice. J Voice. 1998;12:315–327. 10. Andrade PA, Wood G, Ratcliffe P, et al. Electroglottographic study of seven semi-occluded exercises: LaxVox, straw, lip-trill, tongue-trill, humming, hand-over-mouth, and tongue-trill combined with handover-mouth. J Voice. 2014;28:589–595. 11. Kapsner-Smith MR, Hunter EJ, Kirkham K, et al. A randomized controlled trial of two semi-occluded vocal tract voice therapy protocols. J Speech Lang Hear Res. 2015;58:535–549. 12. Morrison MD, Rammage LA. Muscle misuse voice disorders: description and classification. Acta Otolaryngol. 1993;113:428–434. 13. Hillman RE, Holmberg EB, Perkell JS, et al. Phonatory function associated with hyperfunctionally related vocal fold lesions. J Voice. 1990;4:52–63. 14. Roy N, Bless DM, Heisey D, et al. Manual circumlaryngeal therapy for functional dysphonia: an evaluation of short- and long-term treatment outcomes. J Voice. 1997;11:321–331. 15. Van Lierde KM, De Bodt M, Dhaeseleer E, et al. The treatment of muscle tension dysphonia: a comparison of two treatment techniques

16.

17. 18.

19.

20.

21.

22.

23.

24.

25. 26.

27. 28. 29. 30.

31. 32.

33.

34.

35.

36.

9

by means of an objective multiparameter approach. J Voice. 2010;24:294–301. Mathieson L, Hirani SP, Epstein R, et al. Laryngeal manual therapy: a preliminary study to examine its treatment effects in the management of muscle tension dysphonia. J Voice. 2009;23:353–366. Story BH, Titze IR, Hoffman EA. The relationship of vocal tract shape to three voice qualities. J Acoust Soc Am. 2001;109:1651–1667. Shrivastav R, Yamaguchi H, Andrews M. Effects of stimulation techniques on vocal responses: implications for assessment and treatment. J Voice. 2000;14:322–330. Dromey C, Nissen SL, Roy N, et al. Articulatory changes following treatment of muscle tension dysphonia: preliminary acoustic evidence. J Speech Lang Hear Res. 2008;51:196–208. Roy N, Nissen SL, Dromey C, et al. Articulatory changes in muscle tension dysphonia: evidence of vowel space expansion following manual circumlaryngeal therapy. J Commun Disord. 2009;42:124–135. Roy N, Ferguson NA. Formant frequency changes following manual circumlaryngeal therapy for functional dysphonia: evidence of laryngeal lowering? J Med Speech Lang Pathol. 2001;9:169–175. Guzman M, Castro C, Testart A, et al. Laryngeal and pharyngeal activity during semioccluded vocal tract postures in subjects diagnosed with hyperfunctional dysphonia. J Voice. 2013;27:709–716. Mitchell HF, Kenney DT, Ryan M, et al. Defining “open throat” through content analysis of experts' pedagogical practices. Logoped Phoniatr Vocol. 2003;28:167–180. Roy N, Weinrich B, Gray SD, et al. Three treatments for teachers with voice disorders: a randomized clinical trial. J Speech Lang Hear Res. 2003;46:670–688. Titze IR. A theoretical study of F0-F1 interaction with application to resonant speaking and singing voice. J Voice. 2004;18:292–298. Verdolini-Marston K, Burke MK, Lessac A, et al. Preliminary study of two methods of treatment for laryngeal nodules. J Voice. 1995;9:74–85. Stemple JC, Thomas Fry L. Voice Therapy: Clinical Case Studies. San Diego: Plural Publishing; 2010. Rosen CA, Lee AS, Osborne J, et al. Development and validation of the voice handicap index-10. Laryngoscope. 2004;114:1549–1556. Titze IR. The physics of small-amplitude oscillation of the vocal folds. J Acoust Soc Am. 1988;83:1536–1552. Smitheran JR, Hixon TJ. A clinical method for estimating laryngeal airway resistance during vowel production. J Speech Hear Disord. 1981;46:138–146. KayPENTAX. Instruction Manual: Phonatory Aerodynamic System (PAS) Model 6600. 2010. Montvale, NJ. Boersma P, Weenink D (2018). Praat: doing phonetics by computer [Computer program]. Version 6.0.40, retrieved May 2018, from http:// www.praat.org/. Pickett JM. The Acoustics of Speech Communication: Fundamentals, Speech Perception Theory, and Technology. Boston: Allyn & Bacon; 1999. Titze IR, Bergan CC, Hunter EJ, et al. Source and filter adjustments affecting the perception of the vocal qualities twang and yawn. Logoped Phoniatr Vocol. 2003;28:147–155. Laukkanen AM, Lindholm P, Vilkman E, et al. A physiological and acoustic study on voiced bilabial fricative /beta:/as a vocal exercise. J Voice. 1996;10:67–77. Altman KW, Atkinson C, Lazarus C. Current and emerging concepts in muscle tension dysphonia: a 30-month review. J Voice. 2005;19:261– 267.