HHS Public Access Author manuscript Author Manuscript J Voice. Author manuscript; available in PMC 2022 May 01. Published in final edited form as: J Voice. 2021 May ; 35(3): 376–385. doi:10.1016/j.jvoice.2019.09.007. Exercise Science and the Vocalist Aaron M. Johnson1, Mary J. Sandage2 1New York University Voice Center, Department of Otolaryngology – Head and Neck Surgery, New York University School of Medicine, New York, New York 2Department of Communication Disorders, Auburn University, Auburn, Alabama Author Manuscript Abstract The application of exercise science training knowledge has been of growing interest to voice professionals. This tutorial, derived from the authors’ invited presentations from the “Exercise and the Voice” Special Session at the 2018 Voice Foundation Symposium, proposes a foundational theoretical structure based in exercise science, clarifies the wide range of variables that may influence voice training, and summarizes our present understanding of voice physiology from the perspective of muscle training. The body of literature on voice exercise was then analyzed from the perspective of this framework, identifying what we currently know and what we still have yet to learn. Keywords Author Manuscript voice physiology; exercise science; laryngeal muscles; vocal training 1. Introduction Author Manuscript Individual athletes have unique aptitudes and training needs that depend on many physiologic-based or performance-based variables.1 Physiologic variables are related to an individual’s overall physical condition and include aspects such as age, sex, overall health, genetic predisposition, and native muscle fiber type complement. Performance variables are specific to the type of training and include the duration and intensity of the target exercise. These individual differences between athletes are framed in established training paradigms within which power, endurance, or mixed training programs are developed. Occupational voice requirements differ widely between individuals, even within the same profession, e.g., teachers. Some teachers are required to instruct for shorter periods of time at a high sound intensity level in large spaces with acoustic challenges that must be overcome by a combination of voice technique and amplification. Other teachers may speak to smaller Correspondence to Aaron M. Johnson, New York University Voice Center, New York University Langone Health, 345 E. 37th Street, Suite 306, New York, NY 10016, aaron.johnson@nyulangone.org. Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Declarations of interest: The authors have no conflicts of interest to declare. Johnson and Sandage Page 2 Author Manuscript groups in smaller spaces over an eight hour day. Contrasts may also be drawn within vocal performers. The vocal dose and vocal calibration to the room acoustics can vary widely between a recital performance and a lead role in an opera. Despite these individual physiologic and performance differences, few, if any, voice habilitation and rehabilitation programs have been tailored with these differences in mind. Author Manuscript In the field of occupational voice use and training, most clinicians and teachers understand the importance of the specificity of training for targeted skill acquisition that results in excellent technique. Less considered aspects of vocal training are the upregulation of muscle bioenergetics and the development of fatigue resistance to promote injury prevention and more rapid performance recovery. In this tutorial, skill acquisition and upregulation of physiological variables that support extensive muscle engagement will be described within a theoretical framework for training. Frameworks for voice training that include these exercise science constructs of upregulation of muscle cellular function, neuromuscular function, and fatigue resistance, will guide and shape a trajectory for voice training and provide a structure from which empirical studies of voice function can be assessed for strengths and limitations. Evaluating what has been empirically studied to date is as important as identifying what is not yet understood in the merger of exercise physiology and voice training disciplines. 2. Exercise Science Components Author Manuscript Much of what is currently developed for voice training programs has been translated from the evidence available for limb skeletal muscle and cardiorespiratory fitness. There are many aspects of exercise science that play a foundational role in performance training for the end goals of skill acquisition, fatigue management, and avoidance of injury. Knowledge of muscle fiber types, bioenergetic profiles and fatigability of the muscle fiber types in the context of tissue adaptability drives development of sport training programs.2 Skill acquisition has been well addressed in voice training programs; however, a systemic lack of knowledge and awareness of muscle fiber types, bioenergetics, neuromuscular development, and tissue adaptability in the context of occupational voice use undermines our ability to develop the most effective voice habilitation and rehabilitation programs. 2.1. Muscle Fiber Types Author Manuscript The two primary types of skeletal muscle fibers throughout the body are Type I and Type II, differentiated by their contraction speed and fatigability. Type II muscle fibers are the muscle fibers that contribute the most to force production and are more fatigable than Type I muscle fibers. Except for the posterior cricoarytenoid muscle (PCA), the intrinsic laryngeal muscles are generally understood to have more Type II muscle fibers than Type I muscle fibers. The greater prevalence of Type II muscle fibers in the larynx is functionally important for rapid closure of the larynx for airway protection, the primary biological role of the larynx. Type I muscle fibers are endurance fibers, useful primarily for postural support and activity that lasts for longer than 3 minutes. The PCA has a greater complement of Type I muscle fibers than Type II muscle fibers which makes sense given the PCA’s role in maintaining airway patency. J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 3 Author Manuscript The extent to which individuals vary with regard to muscle fiber type complement in the intrinsic laryngeal muscles is not currently well understood.3 It may be that, like limb skeletal muscles, there are individuals who are genetically predisposed to a far larger complement of either Type I or Type II muscle fiber types, providing physiological support for vocal performers we would consider as outliers. Outliers could manifest as either individuals who have little difficulty with extensive use at one extreme or individuals prone to voice disorder at the other extreme. 2.2. Bioenergetic Profiles Author Manuscript Muscle bioenergetics describes the mechanisms and manner by which muscles are supplied with adenosine triphosphate (ATP), the primary fuel for muscle contraction and relaxation. Metabolic pathways include two anaerobic pathways, the immediate energy system and glycolysis, and one aerobic system, oxidative phosphorylation. The energy system is generally inferred by both the length of time one engages in muscle activity and the amount of power produced by the given muscle group.4 While all three metabolic pathways are “turned on” at the start of muscle work, it is generally understood in the limb skeletal muscle literature, that the immediate energy system is depleted first, within the first few seconds of exercise. Glycolysis is the second metabolic pathway to be largely depleted, after about 1-2 minutes. Glycolysis yields about 4 ATP per glucose molecule, making it a rapidly accessible but rather inefficient bioenergetic resource. Athletes who perform ballistic exercise (think hammer thrower) or perform sprint distances are considered to be anaerobic athletes. Author Manuscript After 2-3 minutes of exercise, the aerobic pathway, oxidative phosphorylation, begins to provide an efficient means of ATP production at around 34 ATP per glucose molecule. Therefore, within the temporal framework of muscle metabolism, the anaerobic metabolic pathways are largely depleted within a very short time before the aerobic pathway predominates. Endurance performance bioenergetically translates to continuous exercise that lasts longer than 2-3 minutes, depending on how highly trained the muscle tissue, with faster aerobic production of ATP with a higher fitness level. In actuality, the three mechanisms work synergistically, with the aerobic system predominating after 2-3 minutes of continuous exercise, but glycolysis kicking in should a sudden acceleration of movement be required.4 Author Manuscript There is an emerging body of work that has applied the temporal constructs of bioenergetic substrate utilization to known voicing “on” and “off” intervals in physical education teachers and classroom teachers.5-7 Given that voiced intervals during connected speech in ecologically valid contexts, as determined via voice dosimetry, were overwhelmingly less than 3 seconds in length,5 it has been hypothesized that the intrinsic laryngeal skeletal muscles engaged for phonation are largely reliant on anaerobic metabolism. This observation would be consistent with the limited cadaver muscle fiber typing evidence indicating a higher percentage of Type II muscle fibers in the all of the intrinsic laryngeal skeletal muscles except for the PCA. A primary reliance on anaerobic pathways for occupational voice users would signal a particular type of voice training that would target upregulation, fatigue resistance, and more rapid recovery of the anaerobic energy systems with the specific activity of connected speech. Within the context of accumulated voicing intervals over an entire work day, such as with classroom teaching or call center support J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 4 Author Manuscript agents, it may be hypothesized that a mixed athlete profile (anaerobic + aerobic) would more accurately account for both the rapid acceleration and endurance characteristics of the vocal dose. In this regard, a combined anaerobic/aerobic muscle metabolism would be proposed. Author Manuscript The actual bioenergetic requirements for voice use is largely theoretical at the time of this writing and the metabolic profile for the intrinsic laryngeal muscles is not well understood in vivo. Tellis, et al.8 used visible light spectroscopy in vivo to monitor tissue oxygen and relative total hemoglobin in the thyroarytenoid/lateral cricoarytenoid (TA-LCA) complex. A decrease in oxygen was observed during a sustained voicing task. It is difficult to determine the role of oxygen utilization for production of ATP for this task because glucose uptake was not concurrently measured. Without concurrent measurement of glucose utilization, it would be impossible to determine if laryngeal metabolism was primarily anaerobic or aerobic. Nanjundeswaran, et al.9 used well-established gas exchange analysis to determine oxygen uptake during reading tasks and during recovery from reading in a between group design, individuals describing vocal fatigue, vocally healthy sedentary individuals, and vocally healthy/cardiovascularly conditioned participants. No significant differences in oxygen uptake were identified between the groups. Better understanding of intrinsic laryngeal skeletal muscle metabolism and adaptations with exercise and detraining will be vital for future evidenced-based development of voice habilitation and rehabilitation programs. 2.3. Fatigability/Fatigue Resistance Author Manuscript Fatigue is a complex construct which requires consideration of both psychological and physiological perspectives. In some literature this is described as “central” versus “peripheral” fatigue.10 In other more recent literature, fatigue is framed as “perceived” versus “performance” fatigability.11 Both constructs require consideration of the feeling/ perception of fatigue versus the physiological inability to continue to contract the muscle fibers at the same speed and degree of force as required for the target behavior. To perceive that the voice is tired after having used it for a long time is a different construct from having bioenergetic depletion from extensive engagement of the laryngeal muscles. Quantification of vocal fatigue is an ongoing research effort with voice-related scales such as the Vocal Fatigue Index (VFI) and the Evaluation of the Ability to Sing Easily (EASE) scale now available for use.12,13 Fatigue has generally been considered a negative condition and the remedy has generally been cessation of voice use or voice rest. Author Manuscript There are many idiosyncratic factors that may influence perception of fatigue for voice tasks that include but are not limited to the following: health status, prior training and vocal fitness, self-efficacy, adequate sleep/rest, medication, environment in which the voice tasks takes place, use of amplification, etc. The measurement of voice fatigue may be confounded by fatigue occurring in one or more of the subsystems required for voicing, i.e., respiration, phonation, and articulation/vocal tract tuning. Additionally, vocal fatigue has been attributed to changes in the vocal fold cover resulting from repeated impact forces during voicing, such as increased tissue viscosity and non-muscular tissue strain; however, the evidence to support this has been limited to theoretical modeling and lacks in vivo support.14 In exercise science performance fatigue is not avoided; it is planned for and managed through targeted conditioning programs. When the exercise expectations are known, just as J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 5 Author Manuscript when the vocal dose and the acoustic requirements are known, a training program can be designed to upregulate those physiological processes needed to offset fatigue - what can be referred to as fatigue management. Merging this exercise construct with voice training programs requires a skillful understanding of the end goal for voice production and then adequate time and vocal training for the muscle tissue and motor planning to adapt to the conditions imposed. Muscle tissue upregulates in many aspects - metabolically, neurologically, and morphologically - to be more efficient. The ultimate goal of a fatigue management approach is more rapid recovery from the vocal load imposed and avoidance of injury. To date, there has been little differentiation for occupation-specific voice load and, as a result, many voice professionals are trained in the same way with little thought put toward a fatigue management plan. 3. Neuromuscular Training Adaptations Author Manuscript 3.1. The SAID principle The specific adaptation to imposed demand (SAID) principle is the foundational framework from which any strength and conditioning exercise program is designed for both endurance and ballistic performance goals (Figure 1).15 Muscle tissue adapts to the demands imposed upon it or the lack of demand imposed upon it. This plasticity of muscle tissue is complex and encompasses metabolic, morphologic, and neurologic aspects of muscle tissue function. Author Manuscript A component of the SAID principle, overload, refers to the “imposed demand” aspect of the overarching principle. Overload is achieved when the muscle tissue is challenged with increased intensity and/or duration of use. This challenge may be in the form of increased resistance loading, increased repetition, increased duration of the muscle activity, or a combination of these factors. The exercise principle of specificity, indicates that the muscle will adapt to the activity trained. Specificity is realized via muscle tissue adaptations that encompass motor learning as well. Training for a specific task allows the motor plan to be established so that optimal force and speed of motion are achieved for the target behavior. For example, the velopharyngeal closure differs between speech tasks and other non-speech tasks, such as blowing, sucking, etc. Even within non-nasal consonant targets, the degree of velopharyngeal closure differs, with more forceful and longer closure for the /s/ phoneme than /p/.16 Author Manuscript A final concept to consider within the larger framework of the SAID principle is reversibility. If the muscle is not regularly challenged at the same (maintenance) or greater (overload) degree, then the trained adaptations will downregulate or reverse in order for the muscle tissue to achieve a different level of homeostasis. Applied to voice function, vocal training should be designed in a manner that steadily imposes enough of a demand on the physiology that the individual can achieve the target performance level without overdoing it and risking injury. The type of demand imposed should match the vocal requirements of the goal behavior. For example, preparation for a recital should be different than training for work as a classroom teacher. Analysis of these J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 6 Author Manuscript two different scenarios has indicated that while the overall vocal dose may be similar in distance, the dose is achieved over different time spans and using a different average sound level.17 3.2. Sex & Age Considerations Muscle morphology and response to exercise will differ between most men and women. Women realize measurable muscle strength gains through both neuromuscular mechanisms and hypertrophy, with a greater reliance on the former. Younger men will realize strength gains primarily through muscle hypertrophy. In old age, men have a shift toward neuromuscular mechanisms for strength gains with less reliance on muscle hypertrophy. Lenell, et al.18 provides a comprehensive review of the effects of sex hormones on laryngeal muscle senescence. Author Manuscript 3.3. Training Adaptations in the Muscles of Voice Production Author Manuscript Exercises to increase strength and endurance of the limb muscles have been extensively studied in the traditional exercise physiology literature. Applying these findings to the muscles of voice production presents several challenges. Voice production is a complex, sensorimotor activity that requires the coordination of three major subsystems: the respiratory, phonatory, and resonating/articulatory systems. Many of the respiratory, laryngeal, and cranial muscles that coordinate to produce voice are functionally and microstructurally distinct from the larger limb muscles.19 Little is known about neuromuscular adaptations of these muscles in response to training, partially due to the difficulty in measuring the strength and endurance of these muscles in vivo in humans. Furthermore, the targets of vocal exercises and training are not typically to increase muscle strength or endurance, but to improve vocal function (e.g. increase vocal range, improve vocal quality). Improving vocal function likely depends on changing the muscle activation in multiple subsystems and changing how these subsystems coordinate. Therefore, it is difficult to study specific neuromuscular adaptations underlying changes in vocal function resulting from the simultaneous training of these subsystems. Each of the vocal subsystems has been studied in isolation relative to specific neuromuscular deficits, giving some insight into the capability for training to improve each individual subsystem. Author Manuscript 3.3.1. Respiratory Muscles—In typical voice production there is more than sufficient strength and endurance of the respiratory muscles to produce voice. In voice therapy or singing training, coordination of respiratory muscles with the phonatory system is often the goal rather than muscle strengthening. The capacity for increasing respiratory muscle strength has been demonstrated in respiratory muscle strength training (RMST) programs for individuals who have compromised respiratory muscle strength due to a neurological injury, such as stroke.20 RMST targets either the inspiratory or expiratory muscles using a valved device to introduce a resistance during inspiration or exhalation. Direct measurement of the strength of the respiratory muscles is not possible in vivo. Therefore, respiratory muscle strength is indirectly quantified by measuring changes in pulmonary function parameters such as peak expiratory flow rate and maximum expiratory or inspiratory pressures. This is a reasonable approach given that increased respiratory muscle engagement during these maneuvers has been shown using electromyography.21 J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 7 Author Manuscript The evidence for building respiratory muscle strength is strong; however, generalization of these proposed strength gains to improvements in voice production remains unknown.20 Analyzing respiratory muscle training through the lens of exercise science training principles indicates a mismatch with regard to the specificity of this training for improved voice function. During the RMST tasks, phonation is not coordinated with the isolated respiratory muscle exercise. Therefore, RMST is a good example of a program that is intended to build strength of respiratory muscles which, as a subsystem of voice, could improve voice production for specific individuals with respiratory muscle weakness but may not be a necessary component of voice habilitation or rehabilitation.22 Author Manuscript More relevant for voice production is improved respiratory muscle endurance to reduce fatigue or improve vocal performance during extended periods of voicing, such as singing or sustained speaking (e.g. teaching, lecturing). Respiratory muscle endurance can be trained in patients with conditions that restrict respiration, such as chronic pulmonary obstructive disease.23,24 In these cases, respiratory muscle endurance was measured by calculating maximum voluntary ventilation (MVV), a pulmonary function task. MVV is the total volume of air a person can inhale and exhale over a certain amount of time, typically 12 seconds. In voicing, increasing the duration of sustained exhalation may be of more interest than increasing ventilatory capacity. Although maximum phonation time (MPT) has been used extensively as a surrogate measure of laryngeal aerodynamic efficiency, no specific task targeting respiratory muscle endurance during voicing has been established. Author Manuscript 3.3.2. Muscles of the Vocal Tract—The muscles surrounding the vocal tract contribute to resonance and articulation during voicing by changing the vocal tract’s shape and length. Similar to the respiratory muscles, typical voicing does not likely require greater than normal strength and endurance of the muscles of the vocal tract. Tongue strength during running speech has not been quantified; however, tongue strength measurement during swallowing, a pressure-generating task, has shown that only one-third to one-half of the maximum voluntary tongue strength is engaged during swallowing depending on age.25 Additionally, difficulty with articulatory precision does not necessarily correlate with articulatory muscle weakness. In a study measuring tongue strength in children, there was no difference in tongue strength between children with an idiopathic speech sound disorder and age- and sex-matched peers with typically developing speech.26 However, that same study demonstrated that children with a motor speech disorder did have decreased tongue strength relative to children with either typically developing speech or a speech sound disorder. Author Manuscript The rapid, fine motor muscular movements needed for articulation and resonance adjustments are best trained (and treated when a deficit is present) by speech and voice tasks, not by gross motor, nonspeech exercises.27 In contrast, swallowing is a gross motor function of the vocal tract muscles and, as such, oropharyngeal muscle weakness is assumed to be a significant contributor to oropharyngeal dysphagia. Strengthening exercises for the muscles of the vocal tract, with particular focus on the tongue, have been developed to treat oropharyngeal dysphagia. The tongue is a complex muscular hydrostat, consisting of four intrinsic muscles and five extrinsic muscles.28 It is the most mobile and versatile muscular structure that determines J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 8 Author Manuscript the size and shape of the vocal tract. It is also the primary component for articulatory postures and formant tuning, responsible for producing the formant locations of vowels and the constrictions necessary for most consonants other than bilabial, labiodental, and glottal consonants. Studies of tongue strengthening exercises in the context of dysphagia rehabilitation have shown it is possible to strengthen the tongue muscles with training.29 Tongue strength can be assessed using devices that measure the pressure generated by pushing different portions of the tongue against a pressure transducer, such as the Iowa Oral Performance Instrument and the Madison Oral Strengthening Therapeutic device. 30,31 These devices have also been used to provide feedback during tongue strengthening programs for patients with dysphagia. Author Manuscript Similar to measuring muscle strength in RMST, these tongue devices measure performance of a group of muscles and do not provide direct strength measurements of individual muscles. Also similar to RMST, these tongue training exercises are not likely useful in vocal training unless an individual has known tongue muscle weakness. Presently, the neuromuscular adaptations to tongue strengthening exercises are unable to be directly assessed in humans. Author Manuscript Interestingly, there is evidence that endurance and strength of the tongue are correlated with particular types of physical activity. For example, in a study of tongue strength in different types of athletes, it was shown that weightlifters had greater tongue strength than did runners, while runners had greater tongue endurance than weightlifters.32 It may be that endurance athletes, such as runners, have an innate muscle fiber type composition throughout their body that predicates their affinity for endurance exercises, while the contrary may be true for athletes who excel at ballistic activities. Alternatively, muscle fiber type composition in the tongue may shift with specific types of training. For example, weightlifters often rely on the Valsalva maneuver which would engage the tongue musculature and, theoretically, increase its strength. Further study is needed to understand these potential relationships between of general physical exercise and the muscles voice production. Author Manuscript 3.3.3. Laryngeal Muscles—Sustaining a vowel at a constant pitch and loudness is a task often used to clinically evaluate vocal fold vibration, either through direct observation via endoscopy or indirectly through auditory-perceptual or acoustic analysis. During this task the laryngeal muscles adduct the vocal folds and maintain a constant longitudinal tension. During more typical use, such as conversational speech, the intrinsic laryngeal muscles rapidly adduct and abduct the vocal folds to create voiced and voiceless phonemes and continuously adjust vocal fold tension to change pitch. The very fast speed and small size of the laryngeal muscles makes them phenotypically unique from the larger limb muscles. They have more in common with other muscles in the body that require fast, fine motor adjustments, such as the extraocular eye muscles that rapidly adjust the position of the eye or the muscles that precisely move the fingers to allow us to type or play the piano. Therefore, it is difficult to confidently apply exercise physiology knowledge from the limb muscles to the laryngeal muscles. J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 9 Author Manuscript As with the muscles of respiration and the vocal tract, specifically targeting strength and endurance of the laryngeal muscles is not typically a target of vocal training. However, lack of laryngeal muscle strength and tonicity are relevant concepts in the context of hypofunctional voice disorders. For example, laryngeal muscle atrophy (loss of size and strength) is thought to be the primary cause of presbyphonia, a voice disorder related to advanced age. The evidence for behavioral interventions to treat presbyphonia is emerging but promising.33 Three approaches that have yielded promising results are Vocal Function Exercises, Phonation Resistance Training Exercise, and the Lee Silverman Voice Treatment. 34-36 Each of these three interventions incorporates one or more maximal vocal function task (i.e. maximum phonation time, maximum frequency, or maximum intensity). These maximum function tasks do not likely require maximum activation of the laryngeal muscles, as would a maximum strength exercise applied to the limb muscles. Laryngeal muscle contraction is much greater in non-vocal activities such as swallowing or the Valsalva maneuver. However, these maximum vocal function tasks engage the laryngeal (and likely respiratory) muscles at levels greater than that required for most voicing tasks, which may result in improved function and neuromuscular adaptations via the principles of overload and specificity. As with the respiratory and lingual muscles, these neuromuscular mechanisms are difficult, if not impossible, to study in humans in vivo. Author Manuscript 3.4. Animal Models of Vocal Exercise Author Manuscript Animal models provide direct access to the muscles that are difficult or impossible to investigate in humans in vivo, allowing direct investigation of the neuromuscular mechanisms underlying vocal exercise.37 Additionally, animal models allow us to examine neuromuscular training response in both diseased and normal, non-injured states. Although animal models have long been used to investigate voice physiology, biology and biomechanics, there is limited work using animal models to research neuromuscular response to training or exercise.37 Two key challenges to this type of study are the differences in animal vocalization mechanics and behaviors relative to human vocalizations, and the difficulty in training animals to vocalize in a manner that adequately represents human vocal training. Author Manuscript Investigation of neuromuscular responses to training requires methodology to target the specific muscle(s) of interest. This targeting has been accomplished in exercise models of the animal vocal subsystems using three methods: resistance loading, neuromuscular electrical stimulation (NMES), and behavioral training. Resistance loading increases the amount of work a muscle or set of muscles must do to achieve a functional result. For example, muscle hypertrophy can be induced in the limbs through compensatory overload in one muscle after denervating or otherwise ablating a synergistic muscle. NMES simulates exercise by electrically stimulating the target muscles without the need for behavioral training. The stimulation is applied directly to the nerve supplying the target muscles which consequently mimics voluntary muscle contraction, absent the motor planning component that would be present in volitional muscle engagement. Behavioral training uses operant conditioning to train animals to produce a target behavior for a reward. The target behavior activates the muscles of interest and can be trained to progressively increase engagement of these muscles over time. Rewards in these behavioral paradigms are typically food or water J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 10 Author Manuscript and require the animals be motivated for the reward. Examples of these approaches can be found in each vocal subsystem. Author Manuscript 3.4.1. Animal models of respiratory muscle exercise—Resistance loading has been used to examine respiratory muscle exercise in both rat and sheep models.38,39 In contrast to human studies, in which airway resistance is introduced using an external device, in an animal model resistive loading to the inspiratory or expiratory muscles can be achieved by creating an upper airway obstruction with a tracheal cannula. This obstruction increases the respiratory muscle engagement needed to maintain ventilation. Results using this model demonstrated an interesting mix of adaptations consistent with both strength training, such as increased mass of the diaphragm (muscle hypertrophy), and endurance training, as evidenced by an increase in GLUT-4 protein and the proportion of Type I (fatigue resistant) fibers. Therefore, the neuromuscular response of the respiratory muscles to exercise appears to be a hybrid response of the traditional categories defined in the limb muscles, highlighting the need for direct study of individual muscles and muscle groups within the vocal subsystems rather than extrapolating findings from the limb muscles. 3.4.2. Animal models of tongue muscle exercise—NMES of the tongue muscles via stimulation of the hypoglossal nerve activates all of the intrinsic and extrinsic tongue muscles except for the palatoglossus muscle. Daily NMES of the rat tongue for 8 weeks resulted in increased fatigue resistance of the extrinsic tongue muscles, greater twitch and tetanic forces, and, in general, a shift to more fatigue-resistant muscle fiber types.40 Additionally, the relationship of the pre/post synaptic morphology in the neuromuscular junctions (NMJ) of the genioglossus muscle changed with NMES, implicating possible improved synaptic efficiency.41 Author Manuscript A caveat to use of NMES as a model of exercise is that it differs from voluntary muscle contraction in two important ways. First, the intensity of nerve stimulation in NMES studies often results in supramaximal contraction of the stimulated muscles (i.e. greater intensity than what is possible with voluntary contraction). Second, NMES lacks the element of motor planning and execution that is a feature of skill acquisition to match muscle fiber engagement to task requirements. Therefore, NMES informs us of the potential neuromuscular response of muscle(s) to activation, but translating that response to human behavior is tempered by these aspects that reduce ecological validity. Author Manuscript The behavioral training model has also been applied to the rat tongue using a resistance licking paradigm in which water restriction was used to motivate rats to lick against a force transducer for a water reward. The amount of licking force necessary for the reward progressively increased over 8 weeks, thereby modeling a progressive resistance exercise program. This paradigm resulted in increased tongue forces and shifts in muscle fiber types in the intrinsic tongue muscles from slow to fast or fast to slow, depending on the muscle region.42,43 Tongue exercise did not increase muscle fiber size in the intrinsic tongue muscles, but did result in a trend toward larger cross-sectional muscle fiber area of the genioglossus muscle, the primary protrusive extrinsic tongue muscle used for licking. Although this program trains a protrusive tongue maneuver which is not a typical action in J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 11 Author Manuscript human speech or swallowing, it is a meaningful, functional task for the rat and, therefore, may represent (relative to NMES) a more homologous task to human tongue exercise. Similar to the goals of human investigations of tongue exercise, the motivation for these animal models is to understand mechanisms underlying dysphagia treatments. Although, relative to swallowing, the tongue muscles contract at lower intensities during speech, these basic science investigations help us understand the unique and complex ways in which the tongue muscles respond to increased use. 3.4.3. Animal models of laryngeal muscle exercise—The laryngeal muscles are difficult to study due to their small size, relative inaccessibility, and specialized function. Both NMES and behavioral training have been used to investigate the neuromuscular response of the intrinsic laryngeal muscles to increased use. Author Manuscript In contrast, Karbiener et al.45 demonstrated intrinsic laryngeal muscle hypertrophy in an NMES sheep model. Their NMES protocol stimulated the recurrent laryngeal nerve for only 2 minutes each day for an overall duration of 29 days. Instead of using an intensity that resulted in supramaximal contraction of the muscles, they stimulated the muscles with enough intensity to fatigue the adductory muscles, as quantified functionally. Using a balloon catheter to measure the laryngeal adductory pressure, they experimentally determined the NMES parameters (intensity, duration, etc.) that resulted in muscle fatigue, as measured by decreased adductory pressure over time. This combination, 4 weeks of daily sessions of short sets (2 minute) of muscle contractions leading to fatigue, is consistent with the parameters of strength training paradigms in the limb muscles and, therefore, the resulting hypertrophy in the TA is an expected neuromuscular adaptation. This adaptation is consistent with the hypertrophy observed in Type II skeletal muscles that occurs secondary to shorter duration muscle efforts.3 Author Manuscript Author Manuscript Using an NMES rat model, McMullen, et al.44 stimulated the recurrent laryngeal nerve at a supramaximal level for 2 hours/day for 1 or 2 weeks. Following either dose of NMES, the TA muscle did not hypertrophy but changes in NMJ morphology and mitochondria were consistent with endurance training. These results could be explained by either the duration or the intensity of the NMES protocol. Early (within the first month) neuromuscular adaptations to exercise are typically neural, not hypertrophic, whereas the duration of NMES was either 1 or 2 weeks. The intensity of the NMES was supramaximal, which may have not allowed the muscles to fully recover between sessions, thereby forcing the muscles to become more reliant on aerobic pathways. The length of time the muscle contraction was activated was aligned with reliance on aerobic metabolism which is not typically associated with muscle hypertrophy.3 NMES protocols for stimulating laryngeal muscles using either supramaximal or functional fatigue offer controlled ways of measuring muscle contraction. In contrast, behavioral models offer less precise control over individual muscle contractions but are more homologous to human vocal training. The only behavioral animal model of vocal training currently used to investigate the laryngeal neuromuscular adaptations to vocal exercise is training rats to increase their production of ultrasonic vocalizations (USVs). Despite the J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 12 Author Manuscript different final acoustic output between human vocalizations and rat USVs, the laryngeal muscle activity during the production and modulation of each type of vocalization is quite similar. Laryngeal EMG in awake, vocalizing rats, showed that both the thyroarytenoid and cricothyroid muscles are active during USV frequency modulation, demonstrating that both human and rat USVs require fine motor control of the laryngeal musculature.46 An important difference between human vocalizations and USVs is that USVs appear to be produced via an edge-tone whistle mechanism created by an air jet through a narrowed glottis, not vocal fold vibration.47 This mechanistic difference is actually an advantage of this model; training production of USVs can increase activation of the intrinsic laryngeal muscles regulating vocal fold adduction and elongation without concern for inducing phonotrauma in the superficial vocal fold layers. Author Manuscript Author Manuscript There are two studies using this model that have demonstrated neuromuscular adaptations within the TA muscle of both the young and old rat larynx in response to vocal training.48,49 After 8 weeks of daily vocal training with progressively increasing the target number of USVs each week, age-related changes in neuromuscular junction (NMJ) morphology were reduced. This was the first evidence from a behavioral animal model that vocal exercise could ameliorate age-related changes in laryngeal neuromuscular mechanisms. In this study the behavioral training target (number of USVs produced per session) was the same across all animals, regardless of individual performance or age. In the next study using this model, the training goals were individualized for each animal, resulting in much higher targets.49 Both the NMJ and muscle fiber size were examined in the TA of young rats after 4 or 8 weeks of training. Similar to the first study, adaptations of NMJ morphology were seen after only 8 weeks, but not 4 weeks. No muscle hypertrophy was observed at either time point. It is likely that the functional loads created by USV training are, therefore, inadequate to induce hypertrophy, which is in contrast to the maximal contractions induced by NMES. Further study of behavioral USV training is required to understand the functional consequences of the changes in NMJ morphology. 4. Model of muscular activation during vocalization Author Manuscript In its broadest sense, the goal of vocal exercise is to improve the coordination of the three vocal subsystems (respiration, phonation, and resonance), whether the context is voice therapy for dysphonic individuals or vocal training for vocal athletes. As such, vocal exercise is ultimately fine motor skill training. This hypothesis begs the question, are the concepts of muscle strength or fatigue resistance relevant to vocal training? That is, do any of the considerations outlined in this exercise science tutorial actually matter when it comes to training the voice? In typical day-to-day voice use, they probably do not. Successful vocalization for typical speech communication does not require greater than homeostatic engagement of the respiratory, laryngeal, or articulatory/resonatory muscles. However, in a compromised system (e.g. stroke, muscle tension dysphonia) or during athletic voice use (e.g. singing, preaching, teaching), the muscles of voice production require training and, therefore, the principles of exercise science are relevant and critical to consider. Consider the SAID principle and the theoretical construct presented in Figure 1. Where does vocalization fall in the continuum of endurance to strength training? Strength training J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 13 Author Manuscript focuses on strengthening a specific muscle or groups of muscles. As discussed above, it is difficult to isolate the muscles of voice production. Instead, vocalization can be thought of as a coordinated, repetitive sequence of muscle activation and deactivation, similar to a mixed athlete that requires both endurance and ballistic components. Cyclists require endurance for rides longer than 3 minutes and also require sudden bursts of strength to make it up a hill. Vocalists may also require both endurance and sudden changes in acceleration to meet occupational vocal demands. Author Manuscript Figure 2 presents a model of the cyclical coordination of muscle activation in each vocal subsystem required to produce (A) a sentence commonly used in voice evaluation and (B) a sustained sung phrase. Three phases of vocalization are delineated: inhalation, pre-phonatory setup, and phonation. The figure represents the sequence of the phases, whereas the precise timing of the phases can vary greatly depending on the communicative context and individual differences. For example, the pre-phonatory setup phase is generally quite short, but is presented in Figure 2 largely enough to ease visual representation. Additionally, the figure is meant to be an overview of the coordination of the vocal subsystems, not a detailing of all muscles involved in speech and voice. For example, for review of the complexity of passive and active respiratory forces involved in speech and singing, please see the seminal work by Hixon, Hoit and colleagues.50-52 Author Manuscript To summarize the cycle: The muscles within and across each subsystem must 1) coordinate to inhale a sufficient amount of air for the planned utterance, 2) set the initial laryngeal and articulatory position to begin the utterance, 3) rapidly adjust throughout the utterance to meet the respiratory, phonatory, and articulatory demands of each phoneme while modulating the pitch and intensity of the voice for the appropriate prosody of the intended message. The cycle repeats with each new utterance, sometimes immediately, such as in a lecture or performance situation, or with a delay, as during turn-taking in a conversation. This repetitive, low-load cycle during typical conversational speech does not tax the muscles to the point of fatigue. If, however, the vocal load (intensity or duration) is increased, the neuromuscular load also increases and neuromuscular fatigue may ensue. Additionally, if muscle strength or endurance is comprised due to illness or disease, the cycle may be disrupted or other muscles may engage in an attempt to compensate, thus altering the cycle and possibly disrupting communication. In either situation (increased vocal load or comprised muscle function), vocal training based on the principles of exercise science can be used to restore or habilitate the desired vocal output. 5. Future Directions Author Manuscript To date, the application of exercise science physiology and training principles to vocal function has been pursued with great enthusiasm; however, many of our current beliefs about voice function and vocal training lack evidence. From this tutorial, it is clear that translation of the evidence from limb skeletal muscle physiology is insufficient for our understanding of the muscles used for vocal function. The frameworks that are well developed in exercise science may serve as a starting place to develop investigations targeting both basic and applied aspects of voice physiology. Future investigations should J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 14 Author Manuscript pursue a thorough understanding of intrinsic laryngeal skeletal muscle fiber typing and metabolism, neuromuscular responses to voice exercise within the contexts of age and sex differences, and laryngeal muscle response to extensive voice rest and reduced use. Identifying the physiological, neurological, and psychological aspects of vocal loading and voice fatigue in models that mimic human vocalizations loads will provide the foundation needed for applied research. Applied research will be beneficial to create a framework from which voice habilitation and rehabilitation programs may be developed and evaluated for efficacy. Acknowledgements The authors would like to thank The Voice Foundation and Nancy Solomon, Ph.D. for the invitation to submit this work, based on the authors’ invited lectures on Exercise and the Voice at The Voice Foundation’s 47th Annual Symposium: Care of the Professional Voice on June 1, 2018, in Philadelphia, Pennsylvania. Author Manuscript This work was partially supported by funding from The National Institute on Deafness and Other Communication Disorders (NIDCD/NIH), grant K23DC014517 (Johnson, PI). 7. References Author Manuscript Author Manuscript 1. McArdle WD, Katch FI, Katch VL. Essentials of exercise physiology. Lippincott Williams & Wilkins; 2006. 2. Bompa T, Haff G. Periodization: theory and methodology of training: Human Kinetics. ISBN-13. 2009:978–0736074834. 3. Sandage MJ, Smith AG. Muscle Bioenergetic Considerations for Intrinsic Laryngeal Skeletal Muscle Physiology. J Speech Lang Hear Res. 2017;60(5):1254–1263. [PubMed: 28505224] 4. Brooks GA, Fahey TD, Baldwin KM. Exercise physiology : human bioenergetics and its applications. 4th ed. Boston: McGraw-Hill; 2005. 5. Smith AG, Sandage MJ, Pascoe DD, Plexico LW, Lima IR, Cao G. Elementary School Teachers' Vocal Dose: Muscle Bioenergetics and Training Implications. J Speech Lang Hear Res. 2017;60(7):1831–1842. [PubMed: 28614843] 6. Morrow SL, Connor NP. Comparison of voice-use profiles between elementary classroom and music teachers. J Voice. 2011;25(3):367–372. [PubMed: 20359861] 7. Titze IR, Hunter EJ, Svec JG. Voicing and silence periods in daily and weekly vocalizations of teachers. J Acoust Soc Am. 2007;121(1):469–478. [PubMed: 17297801] 8. Tellis CM, Rosen CA, Carroll TL, Fierro M, Sciote JJ. In vivo oxygen consumption and hemoglobin levels in human thyroarytenoid muscle. Laryngoscope. 2011;121(11):2429–2434. [PubMed: 22020893] 9. Nanjundeswaran C, VanSwearingen J, Abbott KV. Metabolic Mechanisms of Vocal Fatigue. J Voice. 2017;31(3):378 e371–378 e311. 10. MacIntosh BR, Gardiner PF, McComas AJ, eds. Skeletal muscle: form and function. Human Kinetics; 2006. 11. Enoka RM, Duchateau J. Translating Fatigue to Human Performance. Med Sci Sports Exerc. 2016;48(11):2228–2238. [PubMed: 27015386] 12. Nanjundeswaran C, Jacobson BH, Gartner-Schmidt J, Verdolini Abbott K. Vocal Fatigue Index (VFI): Development and Validation. J Voice. 2015;29(4):433–440. [PubMed: 25795356] 13. Phyland DJ, Pallant JF, Benninger MS, et al. Development and preliminary validation of the EASE: a tool to measure perceived singing voice function. J Voice. 2013;27(4):454–462. [PubMed: 23583205] 14. Welham NV, Maclagan MA. Vocal fatigue: current knowledge and future directions. J Voice. 2003;17(1):21–30. [PubMed: 12705816] 15. Powers S Exercise physiology : theory and application to fitness and performance. 9th ed. Boston: McGraw-Hill; 2014. J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 15 Author Manuscript Author Manuscript Author Manuscript Author Manuscript 16. Moon JB, Kuehn DP. Anatomy and physiology of normal and disordered velopharyngeal function for speech. In: Bzoch KR, ed. Comunicative disorders related to cleft lip and palate. 5th ed. ed. Austin, TX: Pro-Ed; 1997. 17. Sandage MJ, Hoch M. Training considerations for recital performance: Framing vocal dose in a fatigue resistance training model. Journal of Singing. 2019;75(4):395–400. 18. Lenell C, Sandage MJ, Johnson AM. A Tutorial of the Effects of Sex Hormones on Laryngeal Senescence and Neuromuscular Response to Exercise. J Speech Lang Hear Res. 2019;62(3):602– 610. [PubMed: 30950744] 19. Sciote JJ, Horton MJ, Rowlerson AM, Link J. Specialized cranial muscles: how different are they from limb and abdominal muscles? Cells Tissues Organs. 2003;174(1-2):73–86. [PubMed: 12784043] 20. Menezes KK, Nascimento LR, Ada L, Polese JC, Avelino PR, Teixeira-Salmela LF. Respiratory muscle training increases respiratory muscle strength and reduces respiratory complications after stroke: a systematic review. J Physiother. 2016;62(3):138–144. [PubMed: 27320833] 21. Sapienza CM. Respiratory muscle strength training applications. Curr Opin Otolaryngol Head Neck Surg. 2008;16(3):216–220. [PubMed: 18475074] 22. Desjardins M, Bonilha HS. The Impact of Respiratory Exercises on Voice Outcomes: A Systematic Review of the Literature. J Voice. 2019. 23. Scherer TA, Spengler CM, Owassapian D, Imhof E, Boutellier U. Respiratory muscle endurance training in chronic obstructive pulmonary disease: impact on exercise capacity, dyspnea, and quality of life. Am J Respir Crit Care Med. 2000;162(5):1709–1714. [PubMed: 11069801] 24. Mador MJ, Deniz O, Aggarwal A, Shaffer M, Kufel TJ, Spengler CM. Effect of respiratory muscle endurance training in patients with COPD undergoing pulmonary rehabilitation. Chest. 2005;128(3):1216–1224. [PubMed: 16162709] 25. Park JS, Oh DH, Chang M. Comparison of maximal tongue strength and tongue strength used during swallowing in relation to age in healthy adults. J Phys Ther Sci. 2016;28(2):442–445. [PubMed: 27064477] 26. Potter NL, Nievergelt Y, VanDam M. Tongue Strength in Children With and Without Speech Sound Disorders. Am J Speech Lang Pathol. 2019;28(2):612–622. [PubMed: 31136240] 27. Weismer G Philosophy of research in motor speech disorders. Clin Linguist Phon. 2006;20(5):315– 349. [PubMed: 16728332] 28. Bunton K, Huber JE. Articulation and Resonance. In: Rousseau B, Branski RC, eds. Anatomy and physiology of speech and hearing. New York: Thieme; 2018:391–442. 29. Adams V, Mathisen B, Baines S, Lazarus C, Callister R. A systematic review and meta-analysis of measurements of tongue and hand strength and endurance using the Iowa Oral Performance Instrument (IOPI). Dysphagia. 2013;28(3):350–369. [PubMed: 23468283] 30. Yeates EM, Molfenter SM, Steele CM. Improvements in tongue strength and pressure-generation precision following a tongue-pressure training protocol in older individuals with dysphagia: three case reports. Clin Interv Aging. 2008;3(4):735–747. [PubMed: 19281066] 31. Rogus-Pulia N, Rusche N, Hind JA, et al. Effects of Device-Facilitated Isometric Progressive Resistance Oropharyngeal Therapy on Swallowing and Health-Related Outcomes in Older Adults with Dysphagia. J Am Geriatr Soc. 2016;64(2):417–424. [PubMed: 26804715] 32. VanRavenhorst-Bell HA, Coufal KL, Patterson JA, Mefferd AS. A comparative study: tongue muscle performance in weightlifters and runners. Physiol Rep. 2018;6(22):e13923. [PubMed: 30460755] 33. Oates JM. Treatment of dysphonia in older people: the role of the speech therapist. Curr Opin Otolaryngol Head Neck Surg. 2014;22(6):477–486. [PubMed: 25250623] 34. Lu FL, Presley S, Lammers B. Efficacy of intensive phonatory-respiratory treatment (LSVT) for presbyphonia: two case reports. J Voice. 2013;27(6):786 e711–723. 35. Sauder C, Roy N, Tanner K, Houtz DR, Smith ME. Vocal function exercises for presbylaryngis: a multidimensional assessment of treatment outcomes. Ann Otol Rhinol Laryngol. 2010;119(7):460–467. [PubMed: 20734967] J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 16 Author Manuscript Author Manuscript Author Manuscript 36. Ziegler A, Verdolini Abbott K, Johns M, Klein A, Hapner ER. Preliminary data on two voice therapy interventions in the treatment of presbyphonia. Laryngoscope. 2014;124(8):1869–1876. [PubMed: 24375313] 37. Johnson AM. Basic Science: The Foundation of Evidence-Based Voice Therapy. Perspectives of the ASHA Special Interest Groups. 2016;1(3):7–13. 38. Tarasiuk A, Scharf SM, Miller MJ. Effect of chronic resistive loading on inspiratory muscles in rats. J Appl Physiol (1985). 1991;70(1):216–222. [PubMed: 2010379] 39. Akabas SR, Bazzy AR, DiMauro S, Haddad GG. Metabolic and functional adaptation of the diaphragm to training with resistive loads. J Appl Physiol (1985). 1989;66(2):529–535. [PubMed: 2540138] 40. Connor NP, Russell JA, Jackson MA, et al. Tongue muscle plasticity following hypoglossal nerve stimulation in aged rats. Muscle Nerve. 2013;47(2):230–240. [PubMed: 23169566] 41. Johnson AM, Connor NP. Effects of electrical stimulation on neuromuscular junction morphology in the aging rat tongue. Muscle Nerve. 2011;43(2):203–211. [PubMed: 21254085] 42. Connor NP, Russell JA, Wang H, Jackson MA, Mann L, Kluender K. Effect of tongue exercise on protrusive force and muscle fiber area in aging rats. J Speech Lang Hear Res. 2009;52(3):732–744. [PubMed: 18723593] 43. Cullins MJ, Krekeler BN, Connor NP. Differential impact of tongue exercise on intrinsic lingual muscles. Laryngoscope. 2018;128(10):2245–2251. [PubMed: 29243257] 44. McMullen CA, Butterfield TA, Dietrich M, et al. Chronic stimulation-induced changes in the rodent thyroarytenoid muscle. J Speech Lang Hear Res. 2011;54(3): 845–853. [PubMed: 21106694] 45. Karbiener M, Jarvis JC, Perkins JD, et al. Reversing Age Related Changes of the Laryngeal Muscles by Chronic Electrostimulation of the Recurrent Laryngeal Nerve. PLoS One. 2016;11(11):e0167367. [PubMed: 27893858] 46. Riede T Subglottal pressure, tracheal airflow, and intrinsic laryngeal muscle activity during rat ultrasound vocalization. J Neurophysiol. 2011;106(5):2580–2592. [PubMed: 21832032] 47. Riede T, Borgard HL, Pasch B. Laryngeal airway reconstruction indicates that rodent ultrasonic vocalizations are produced by an edge-tone mechanism. R Soc Open Sci. 2017;4(11):170976. [PubMed: 29291091] 48. Johnson AM, Ciucci MR, Connor NP. Vocal training mitigates age-related changes within the vocal mechanism in old rats. J Gerontol A Biol Sci Med Sci. 2013;68(12):1458–1468. [PubMed: 23671289] 49. Lenell C, Newkirk B, Johnson AM. Laryngeal Neuromuscular Response to Short- and Long-Term Vocalization Training in Young Male Rats. J Speech Lang Hear Res. 2019;62(2):247–256. [PubMed: 30950702] 50. Hixon TJ, Goldman MD, Mead J. Kinematics of the chest wall during speech production: volume displacements of the rib cage, abdomen, and lung. J Speech Hear Res. 1973;16(1):78–115. [PubMed: 4267384] 51. Watson PJ, Hixon TJ. Respiratory Kinematics in Classical (Opera) Singers. J Speech Hear Res. 1985;28(1):104–122. [PubMed: 3157022] 52. Hoit JD, Plassman BL, Lansing RW, Hixon TJ. Abdominal muscle activity during speech production. J Appl Physiol (1985). 1988;65(6):2656–2664. [PubMed: 3215865] Author Manuscript J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 17 Author Manuscript Figure 1. Exercise science theoretical framework. Author Manuscript Author Manuscript Author Manuscript J Voice. Author manuscript; available in PMC 2022 May 01. Johnson and Sandage Page 18 Author Manuscript Author Manuscript Author Manuscript Figure 2. Model of neuromuscular coordination during vocalization of (A) a spoken utterance and (B) a sustained sung phrase. Author Manuscript J Voice. Author manuscript; available in PMC 2022 May 01.