Potential Biophysiological Mechanisms Underlying Vocal Demands and Vocal Fatigue

Summary:

Patients with complaint of vocal fatigue have perceptual, acoustic, and aerodynamic outcomes that are heterogeneous in nature. One reason may be due to different underlying biophysiological mechanisms that lead to these heterogeneous clinical presentations. Five potential mechanisms are proposed: neuromuscular, metabolic, vocal tissue, afferent, and central neural. Analytical frameworks and study designs to study these mechanisms are also addressed. A better understanding of biophysiological mechanisms of vocal fatigue can improve precision of therapeutic approaches. It can also help shift management from symptom-based to etiology-focused approaches for vocal fatigue.

INTRODUCTION

Vocal fatigue has traditionally been characterized by patient perception of symptoms, including vocal effort, strain, tiredness, weakness, and changes in vocal quality—to name a few.^1–3 Attempts have also been made to characterize vocal fatigue by clinical metrics (perceptual, acoustic, and aerodynamic).^4,5 However, symptoms and clinical metrics vary across individuals with complaint of vocal fatigue. For example, some individuals may notice changes in their vocal quality with onset of vocal fatigue, which may lead to corollary changes in acoustic vocal output.^6,7 Others have no changes in perceptual or acoustic vocal parameters, but do notice increased physiological effort or strain to produce their voice with vocal fatigue onset.⁸

Several theoretical frameworks have been proposed to better understand vocal fatigue. For example, Hunter et al, 2020 conceptualize vocal fatigue in the context of its vocal demands (load) and vocal demand-responses (loading). They define vocal demands as the vocal requirements for a given communication scenario, independent of the individual’s physiology, techniques, and perception.⁹ These vocal demands may involve different ways in which the individual uses their voice (eg, changes in vocal loudness, pitch, duration, speech rate, and quality) which can also be influenced by external factors like room size and ambient noise levels. Hunter et al, also define vocal demand-responses as the reactive voice productions to specific communication scenarios.⁹ Different vocal behaviors may lead to different vocal demand responses and can be influenced by various underlying biophysiological mechanisms (eg, although hypothetical, one example is increased vocal intensity or chronicity of vocal demands could lead to changes in breathing patterns, which over time, can result in vocal fatigue from metabolic changes in the respiratory or phonatory system).

Vocal demands can be acute or recurrent. For example, individuals may experience acute vocal fatigue after one session of heavy vocal demands that resolve with adequate rest and recovery. Conversely, vocal demands that are recurrent occur over multiple weeks, months, or years. Vocal demand responses may be adaptive or maladaptive. An adaptive response may be one that is able to recover from the demand placed on the system. This type of vocal demand response is common in occupational voice users with high vocal demands. In contrast to adaptive responses, a maladaptive response is one in which the system is unable to adapt to meet increased acute or recurrent vocal demands. Debilitating maladaptive responses—typically to recurrent vocal demands—may result in individuals seeking medical care or behavioral management for their vocal fatigue (especially those in vocally demanding vocations).

Why some vocal systems adapt while others maladapt may have to do with different biophysiological mechanisms underlying symptoms of vocal fatigue. It is important to consider these mechanisms in any clinical phenomenon to optimize precision of treatment beyond superficial or band aid approaches that focus on symptom severity or frequency reduction. Although several mechanisms specific to vocal fatigue have been proposed in the literature—including neuromuscular fatigue, vocal fold tissue cover fatigue, and metabolic fatigue^10–13—the context of these potential biophysiological mechanisms in the clinical realm have largely been ignored.

One reason relationships between clinical symptoms and underlying mechanisms are ignored in the clinical sphere may be that these mechanisms are thought to be solely theoretical and lack direct clinical application. However, a better understanding of these mechanisms and others—especially in the context of clinical presentations—could not only improve management of the condition but could also disentangle variability in clinical symptoms and traditional voice outcome metrics found across studies. For example (and hypothetically speaking), a rough vocal quality in individuals with vocal fatigue could point to vocal fold tissue cover fatigue while a breathy or asthenic vocal quality in patients with vocal fatigue may point towards neuromuscular fatigue. Subglottic pressures may be lower in one type of fatigue (eg, neuromuscular fatigue) but higher in another type of fatigue (eg, vocal cover fatigue). Biophysiological variation across these different types of fatigue may each require different types of medical management or voice therapy approaches. For example, vocal strengthening or improved neuromuscular coordination in the vocal system may be key to neuromuscular fatigue while anti-inflammatory agents or voice rest may be at the core of vocal cover fatigue treatment.

Of course, relationships between vocal output (eg, rough vocal quality) and standard clinical metrics (eg, subglottic pressures, auditory-perceptual) to inference underlying vocal fatigue mechanisms (eg, neuromuscular versus vocal fold cover fatigue) have not been directly studied or compared and are therefore all hypothetical. However, they are worth cogitating. In the following sections, we summarize previous literature on proposed mechanisms specific to the vocal subsystems muscles—neuromuscular fatigue, vocal fold cover tissue fatigue, metabolic fatigue, and central fatigue—using updated sources from the voice and limb exercise physiology literature, where appropriate. We also add an additional potential biophysiological mechanism—afferent fatigue—to the theoretical and analytical framework. Finally, we propose analytical approaches and experimental designs to shift the vocal fatigue framework from hypothetical-based to data-driven knowledge.

POTENTIAL BIOPHYSIOLOGICAL MECHANISMS

Because of the lack of mechanistic studies in the context of human voice biophysiology, we instead turn to literature from other domains to inform our hypotheses. For example, much of what we know about neuromuscular and metabolic mechanisms stem from the limb biophysiology literature.¹⁴ Although muscles of the limbs and larynx differ in their phenotypic expression,^15,16 extrapolation of evidence from the limb biophysiology literature is needed until we have more direct evidence of neuromuscular and metabolic effects of increased vocal demands specific to the intrinsic or extrinsic laryngeal muscles.^14,17 Similarly, the majority of the theories proposed in the afferent and vocal fold cover section rely on animal models to inference clinical applicability. Proposed central neural mechanisms stem from neurobiological imaging in healthy individuals and metabolic studies in animals with experimentally induced limb fatigue.

In the following sections, theoretical differences in healthy adaptation will also be distinguished from maladaptive responses to high vocal demands—in the context of demand chronicity (one type of vocal demand). We stress the importance of delineating adaptation (ie, function) from maladaptation (ie, dysfunction) when studying mechanisms and their relationships to clinical features of vocal fatigue. To understand dysfunction in the context of the vocal mechanism and vocal fatigue, we first need a better foundational understanding for what is normal. Potential experimental designs will also be proposed that could elucidate relationships between biophysiological mechanisms and clinical presentations of vocal fatigue.

Neuromuscular mechanisms

It is estimated that occupational voice users engage their intrinsic laryngeal muscles at least 1800 times an hour for vocal posturing, including cyclical adductory vocal onset and abductory vocal offsets.¹⁸ Therefore, the vocal neuromuscular system likely plays a role in adaptation and maladaptation to increased vocal demands. However, very little data exist on the vocal neuromuscular system in the context of vocal fatigue. Therefore, it is not yet clear whether acute or recurrent engagement of the intrinsic or extrinsic laryngeal muscles can result in neuromuscular vocal fatigue. Of note, although increased neuromuscular demands could also lead to muscle stiffness, discomfort, and pain in the intrinsic or extrinsic laryngeal muscles, elucidation of neuromuscular mechanisms underlying vocal strain is out of the scope of this review.

Neuromuscular adaptations to acute vocal demands

Hypothetically—and based on what we know from the limb muscle literature—adaptive neuromuscular responses to acute vocal demands (one session) may involve increased motor recruitment to generate greater force to meet these demands—for example, to increase vocal intensity or pitch. One study specific to the larynx used hook-wire electromyography in the thyroarytenoid and cricothyroid laryngeal muscle in a small cohort of vocally healthy subjects. Increased contractile force in these laryngeal muscles correlated with brief periods of increased pitch and vocal intensity.¹⁹ Interestingly, greater force activity on electromyography occurred with pitch compared to intensity, demonstrating the need to consider the impact of different vocal behaviors and demands (intensity versus pitch manipulation) on laryngeal neuromuscular vocal demand-responses. One potential future direction involves investigations into intra-subject relationships between intramuscular electromyography, acoustic output, and aerodynamics with systematic manipulations of vocal demand ingredients (eg, intensity, pitch, session duration). Findings would elucidate the effects of different acute demands on the neuromuscular vocal system as they relate to vocal output within and across vocally healthy individuals.

Neuromuscular maladaptation to acute vocal demands

In contrast to neuromuscular adaptations, maladaptive neuromuscular responses to an acute vocal demand may involve insufficient motor activity to meet the demand. One possible iteration—based on what we know from limb muscle biophysiology—would be increased motor unit recruitment and an increase in firing rate but with a decrease in force production. The result would theoretically be weaker contraction (lower electromyographic amplitudes or spectral compression) of vocal muscles from reduced motor unit force. For example, one study used intramuscular electromyography within the intrinsic laryngeal muscles in a small cohort of vocally healthy participants. Results were noticeable shifts in electromyographic spectral compression of the lateral cricoarytenoid laryngeal muscles (as well as the cricothyroid and thyroarytenoid muscle in some of the participants) with increased vocal intensity over an acute period of time. These findings suggest a decrease in neuromuscular force can occur with acute increases in vocal demands—especially vocal intensity—when the demands exceed capacity and lead to acute vocal fatigue (even in vocally healthy individuals).²⁰ Combining intramuscular electromyography with standard acoustic and aerodynamic measures in patients with vocal fatigue, compared with healthy voice users, could elucidate key variable differences within individuals and between groups with an acute vocal demand task.

Neuromuscular adaptations to recurrent vocal demands

Adaptive neuromuscular responses to recurrent demands likely involve morphological changes within the laryngeal neuromuscular system (eg, motor units) to meet these reoccurring demands. Due to methodological and ethical constraints surrounding the study of vocal muscle in humans, we instead turn to animal models to inform how the laryngeal muscles adapt to increased vocal demands. Several studies using a rat model across 4–8 weeks of ultrasonic vocalization training found significant morphological differences in the neuromuscular junction of the thyroarytenoid laryngeal muscle between trained and untrained animals, ie, more compact motor endplates in trained animals.^21,22 These morphological changes are thought to result in improved connectivity between the nerve and laryngeal muscle and may increase synchrony across motor units—leading to greater neuromuscular efficiency. These changes correlated with increased intensity, duration, and complexity of rat ultrasonic vocalizations, suggesting relationships between neuromuscular adaptive changes and acoustic output. Similar studies could also be conducted in the future using subglottic pressure measurements in rats to determine relationships between neuromuscular adaptations and aerodynamics. This work was previously conducted on acute vocal demands but has not been investigated in recurrent increased rate of vocal demands.²³ Additional extrapolation on literature from limb physiology on long-term training and neuromuscular adaptations has been recommended in the context of clinical applicability in the voice therapy room, albeit the information is mostly hypothetical.¹⁴

Neuromuscular maladaptation to recurrent vocal demands

Maladaptive neuromuscular responses to recurrent demands are much less well understood due to lack of empirical data. One hypothesis is that these maladaptive responses involve chronic motor unit recruitment decrements in the primary muscles of vocalization (eg, abductory and adductory intrinsic laryngeal muscles) and respiratory muscles. Although at this point speculative due to lack of data, over time, these neuromuscular patterns could result in increased motor unit recruitment of accessory muscles—like the extrinsic laryngeal and base of tongue muscles—as a compensatory strategy. The result would theoretically be reductions in efficient muscle patterns, longer recovery times, and persistent decline in endurance capabilities in the accessory muscles. This phenomenon has been observed in upper limb and shoulder overuse injuries but has not been elucidated in the vocal system.²⁴ However, one potential way to test this hypothesis is with the use of surface electromyography to measure muscle activation or ultrasound elastography to measure muscle stiffness in the extrinsic accessory muscles—comparing individuals with and without complaint of chronic, debilitating vocal fatigue. These methods can also be paired with laryngoscopy or intrinsic intramuscular laryngeal electromyography to study relationships between the intrinsic laryngeal muscle patterns and extrinsic laryngeal muscle activation.

Metabolic mechanisms

Several theoretical papers have suggested the role of metabolic mechanisms on vocal system adaptation and maladaptation.^17,25,26 However, very little is known about these metabolic responses to acute and recurrent demands in the context of typical voice biophysiology and vocal fatigue. In spite of this gap, recent studies on the effects of increased demands on laryngeal muscle structure in animal models and well-established limb literature can provide some theoretical insight. Studies on systemic bioenergetic utilization with increased vocal demands in humans is also emerging. These studies will be described below.

Metabolic adaptations to acute vocal demands

There is theoretical debate on the types of adaptive bioenergetic pathways utilized by the laryngeal muscles to meet acute vocal demands, but little direct evidence to support or refute these theories. Some have postulated increased reliance on anaerobic substrates with increased vocal demands because of the fast vocal onsets and offsets that last less than 2 minutes during vocal speech communication.^17,25 Others have suggested the cyclical on-and-off cycles of vocal onsets and offsets during communicative vocalizations are more reminiscent of repetitive endurance exercise like running or cycling—which depend largely on aerobic metabolism to meet increased acute demands.^22,26 Recent work suggests adaptive responses to acute demands may be more aerobic than anaerobic in nature. For example, Nanjundeswaran and colleagues found cardiovascularly fit, vocally healthy individuals reached a plateau in oxygen consumption within the first 3 minutes of vocal load and recovered quickly postacute vocal demand, suggesting aerobic adaptation may underlie acute increases in vocal demands in vocally adaptive systems.¹³

Metabolic maladaptation to acute vocal demands

In contrast to metabolic adaptations, maladaptive responses to acute demands may involve greater reliance on anaerobic mechanisms. A greater reliance on anaerobic mechanisms can possibly lead to an accumulation of metabolic byproducts (eg, inorganic phosphate, lactate, hydrogen ions) and extracellular fluid in response to high levels of lactic acid, causing intramuscular edema and erythema.^12,27,28,29 However, much of this information stems from the limb literature and requires direct study in relation to the voice biophysiology. Metabolic profiling of circulating substrates in and around the vocal muscles could help elucidate the specific metabolic substrates involved.

Metabolic adaptations to recurrent vocal demands

Adaptive responses to recurrent vocal demands are not well understood due to difficulties directly accessing the muscles involved in voice production (in vivo) especially over a period of time (longitudinal studies). However, several studies using animal models may provide some insights. Previous work on thyroarytenoid laryngeal muscle in a rat model has shown adaptive morphological changes in the motor endplate suggestive of aerobic (endurance) responses in the muscle-to-nerve connections with recurrent increases in vocal demands in animal models after 4–8 weeks of vocal training.^21,22,26 Additionally, extrapolations from the limb literature on limb muscle strength and endurance training have been proposed to improve metabolic bioenergetic pathways to enhance task performance and recovery in the vocal subsystem muscles, although specifics of these pathways currently lack.¹⁴

Metabolic maladaptation to recurrent vocal demands

Individuals with maladaptive responses to recurrent demands (ie, chronic vocal fatigue) may rely more heavily on anaerobic mechanisms to meet demands. In one study, individuals with chronic vocal fatigue had greater oxygen deficits during conversational and loud reading tasks, with increased postload oxygen consumption after the tasks compared to healthy voice users.¹³ These patterns suggest heavier reliance on anaerobic energy in those with chronic vocal fatigue; it could also result in increased need for neuromuscular recruitment and decreased blood flow to constricted muscles, making metabolic byproduct clearance more challenging. Clinical implications of these findings may be that voice therapy approaches in patients with metabolic vocal fatigue may need to include vocal endurance exercises or cardiovascular training to improve aerobic utilization and reduce reliance on anaerobic mechanisms. A recent study found patients with symptoms of chronic vocal fatigue (as measured by Factor 1 on the Vocal Fatigue Index²⁹) improved with cardiovascular conditioning (Nanjundeswaran, unpublished).

Vocal fold cover mechanisms

The vocal fold cover (epithelium and superficial laminal propria) plays an important role in the vibratory characteristics of voice production. Prior work has implicated tissuerelated fatigue in the vocal fold cover as a potential mechanism underlying vocal fatigue.^10,11 Inferences on localized shearing impact stress to the vocal fold cover from oscillatory collisions between the two vocal folds during vocal fold vibrations have been drawn mainly from animal models and computational modeling.^30,31 The frequency at which these vocal fold collisions occur (eg, from increased vocal duration or pitch) and the force at which they occur (eg, from increased vocal intensity) may influence vocal demand responses and could lead to faster onset of vocal cover fatigue in some individuals.

Vocal fold cover adaptations to acute vocal demands

Adaptive responses to acute increases in vibratory vocal demands involve increased blood flow and extracellular fluid that act like a cushion to prevent damage from impact collision stress or mechanical shearing forces to the vocal fold cover. These protective mechanisms can result in clinical signs of acute vocal fold edema that lasts about 24 hours and then resolves with vocal rest and recovery after 72 hours.³²

Vocal fold cover maladaptation to acute vocal demands

In contrast to vocal cover adaptations, maladaptive changes to acute demands in the vocal fold cover may manifest as impact- or shearing-induced epithelial cell death and separation of elastin fibers from the basement membrane.³³ These acute changes may also be due to inertial stress forces that alter fluid distribution within the vocal fold cover.³¹ The theory is that these maladaptive responses occur in response to excessive medial compression, increased underlying laryngeal muscle tone, or increased mucosal amplitude from the oscillating vocal fold cover.^32,34

Vocal fold cover adaptations to recurrent vocal demands

Adaptive responses in the vocal fold cover to recurrent demands may involve chronic pro-inflammatory vocal fold modifications, ongoing tissue repair, or long-term tissue changes in the connective tissue of the lamina propria from repetitive stress that resolve with rest and recovery after multiple bouts of heavy voice use.^32,35

Vocal fold cover maladaptation to recurrent vocal demands

In contrast to vocal fold cover adaptations, maladaptive responses to recurrent demands could look like poor tissue recovery, minimal tissue repair, or even result in phonotraumatic lesions (eg, vocal fold nodules, hemorrhagic polyps).³⁶ These patterns could start as a maladaptive response to acute demands that gets compounded by chronicity of use. Individuals with chronic challenges producing soft voice and maintaining appropriate phonation threshold pressures may point to those with maladaptive responses to increased vocal demands in the vocal fold cover.^32,37–39 Future work using phonation threshold pressures, along with standard acoustic metrics and vocal fatigue severity scales may help further inform relationships between tissue fatigue, acoustic output, and self-perception of vocal fatigue.

Afferent mechanisms

Little is known about the role of sensory receptor adaptations and maladaptation in the vocal muscles during acute and chronic vocal demands in humans. However, considering the well-known relationships between afferent responses from the somatosensory system and self-reported sensations of pain, discomfort, and effort, there is likely a strong relationship between vocal system afference and the perception of vocal fatigue. Much of the work on afferent mechanisms in vocal production stem from stimuli perturbations in animal studies. This work has yielded a better understanding of the role the afferent vocal system plays in sensation and modulation of vocal fold position, movement, and tension during phonation.⁴⁰ However, there is a critical gap in our understanding of the role the afferent vocal system plays in vocal fatigue. Much of what we include in this section are inferences from other bodies of literature to inform hypothesis-driven concepts surrounding vocal adaptations to increased demands, vocal fatigue, and vocal system afference.

Afferent adaptations to acute vocal demand

Increased airflow rates passing through the glottis during acute demands stimulate myotatic mechanoreceptors in the intrinsic laryngeal muscles. The respective adaptive responses of these mechanoreceptors to mechanical disturbances (like changes in vocal oscillations, pressures, or airflows) is to initiate reflexive muscular adjustments to ensure the vocal folds remain at the intended length and tension during brief periods of increased vocal intensity or phonatory duration.⁴¹ However, how these reflexive adjustments relate to onset of vocal fatigue requires elucidation. Considering the afferent vocal system is essential to the sensation of muscle activation, effort, and tension, it likely plays a role in the sensation of vocal fatigue as well. Hypotheses specific to maladaptive vocal responses will be proposed in subsequent sections.

Afferent adaptations to recurrent vocal demands

There is a lack of direct study on the role of adaptive laryngeal afferent responses to recurrent demands. However, several inferences can be made from previous work. The fact that a trained singer has the ability to produce pitches and intensities with increased precision with practice suggests there is adaptive somatosensory monitoring within the laryngeal phonatory system. Previous work has shown that with practice, singers shift from a heavy reliance on the auditory system for error detection to one of unconscious tactile-proprioceptive regulation.⁴² Sundberg (1973) proposed singers rely on laryngeal musculature (ie, somatosensory muscle memory) while untrained singers rely more on their auditory system.⁴²

Afferent maladaptation to acute and recurrent vocal demands

In contrast to afferent adaptations to vocal demands, annulospiral endings within the inferior border of the vocal folds may be involved in maladaptive responses to acute or recurrent demands. These receptors initiate greater reflexive responses within the intrinsic laryngeal muscles.⁴³ With acute demands, aberrant receptor responses may manifest as paroxysmal sensory feedback below typical response thresholds or as amplified receptor responses to specific stimuli. With chronicity, these receptors may get more and more sensitive, leading to what we clinically perceive as laryngeal hypersensitivity—or heightened sensations that lead to amplified perceptions of vocal fatigue (even when the vocal motor system looks structurally or physiologically “normal” on laryngoscopic examination or sounds normal on auditory-perception or acoustic analysis—which is often the case in individuals with self-perceived vocal fatigue).

Inferences on maladaptive afferent responses to recurrent vocal demands can also be made using studies on III/IV afferents in the limb muscles. Muscle III/IV afferents in the limbs are thought to underlie perceptions of muscle effort, fatigue, discomfort, and pain in the limbs during exercise.^44–47 Muscle III/IV afferent responses—and subsequent increased sensations in exercising muscles—can become amplified with chronicity and recurrence of exercise overload. Considering individuals with chronic vocal fatigue commonly experience corollary symptoms similar to those with exercise intolerance—including perceptions of vocal effort, fatigue, discomfort, and odynophonia with increased voice use—studying relationships between laryngeal afferents and perceptual clinical outcomes are a worthwhile pursuit. For example, the study of reflexive thresholds of laryngeal responses combined with effort, discomfort, and fatigue scales may help us learn more about relationships between afferent mechanisms and clinical perceptions of fatigue.

Central neural mechanisms

Little is known about the effects of vocal demands on central neural response mechanisms in the context of vocal system adaptation and maladaptation. Most of the literature is on the effects of cognitive demands (eg, mental effort) and psychological stress (eg, Stroop test, public speaking task) on vocal behaviors (eg, acoustics).^{48,49,50–52}

Central adaptations to acute vocal demands

Although much of the work on central mechanisms centers around cognitive and psychological demands and not vocal demands, previous neuroanatomy and neurophysiology studies on vocal control suggest adaptive responses to acute vocal demands likely occur somewhere within the primary neural pathways that coordinate phonatory muscle groups and generate appropriate vocal patterns for precise vocal production.⁵³ The first pathway—the reticular formation in the pons and medulla—has direct connections to the phonatory muscles. The second pathway is the anterior cingulate cortex and periaqueductal gray in the midbrain—involved in emotional vocalizations. The third pathway is the primary motor cortex as well as the putamen, globus pallidus, pontine gray, and cerebellum—associated with learned vocal behaviors like speech and singing.^54–56 Immediately posterior to the motor cortex is the somatosensory cortex, which governs sensory information for phonation and articulation. The anterior portion of the insula is involved in somatosensory feedback as well as voluntary respiratory control during vocalizations.^57,58 The anterior insular is also involved in interoception (bodily feelings) and emotions,^59,60 and may play a role in self-perception of vocal effort and fatigue.

Central adaptations to recurrent vocal demands

Recurrent increases in vocal demands likely induce adaptive synaptogenesis to meet these demands. There is also the possibility of cortical reorganization with chronic changes in demands requiring refinement of sensorimotor control. Inferences of these adaptive changes can be made from a study using functional magnetic resonance imaging (fMRI) in singers and non-singers. This study demonstrated the importance of the somatosensory-vocal feedback loop in acquisition of singing skills. Greater activity in the somatosensory cortex was observed in singers who practiced singing more often.⁶¹ Singers also sang more accurately than non-singers with anesthesia, which suggests that those with more vocal training (ie, adaptive demands) rely more heavily on internal models of somatosensory representation. These findings could explain why experienced singers can sing in tune in spite of high levels of competing auditory sounds in the background.

Central maladaptation to acute and recurrent vocal demands

Maladaptive neural responses to acute and recurrent increases in vocal demands in vocal fatigue are far less well understood. However, neurobiological studies on exercise fatigue in the limbs can provide some insights into central neural mechanisms involved in vocal fatigue. Incremental decreases in discharge rates of neurons in the primary motor cortex have been observed with repetitive isometric elbow-flexion in monkeys.⁶² Notably, the magnitude of discharge within motor to somatosensory cortex projections is thought to correlate with the perception of physical effort.^63,64 Neuromodulatory alterations may also underlie central neural vocal fatigue. One study found increased serotonin concentrations in the brain can impair central nervous system function during prolonged exercise. Increased concentration of neuromodulators like serotonin are thought to correlate with decreases in exercise performance.⁶⁵ Increases in tryptophan (a precursor to 5-HT serotonin) and inhibition of dopamine have also been seen in conjunction with decreased motivation, increased lethargy, reduced efficiency, and loss of motor drive and coordination.^64–68 It would be interesting to observe discharge amplitudes in the nucleus ambiguous or laryngeal primary motor cortex or identify neuromodulator concentrations with perception scales of vocal effort and fatigue during a vocally demanding task and between individuals with and without chronic vocal fatigue.

See Table 1 for summary of five potential biophysiological mechanisms as they relate to vocal demand chronicity (acute versus recurrent) and adaptation of vocal demand responses (adaptive versus maladaptive).

Table 1.

Summary of Potential Biophysiological Mechanisms Underlying Vocal Fatigue Symptomology. The effects of acute and recurrent vocal demands on adaptive and maladaptive vocal demand-responses. Mechanisms listed are specific to the vocal system. The “unknown” denotes lack of direct studies on voice biophysiology. Hypothetical underlying mechanisms from other areas of literature as a preliminary model—from which to start developing experimental designs—have been provided in the body of the paper but are not listed here

		Vocal Demands Chronicity
		Acute	Recurrent
Vocal Demand-Responses	Adaptive	Neuromuscular: increased motor recruitment in vocal muscles (Lindestad et al, 1991; Riede 2013) Metabolic: unknown, but increased oxygen uptake kinetics suggests greater reliance on aerobic substrates (Nanjundeswaran et al, 2017) Cover: increased blood flow and extracellular fluid to prevent tissue damage (Hunter & Titze, 2009) Afferent: myogenic reflexive responses for optimal vocal length and tension during phonation (Wyke 1974) Central neural: unknown	Neuromuscular: neuromuscular junction morphology changes for improved motor unit recruitment (eg, Shembel et al, 2021, Johnson et al, 2103, Lenell et al, 2019; McMullen et al 2011) Metabolic: decreased fiber area, increased mitochondrial content and glycogen-positive fibers in laryngeal muscle (McMullen et al 2011); improved systemic oxygen uptake kinetics (Nanjundeswaran et al, 2017) Cover: chronic inflammation (edema); changes in connective tissue that resolves with rest and recovery (tissue repair) (Hunter et al, 2009; Verdolini et al, 2003) Afferent: unknown, but may involve improvements in tactile-proprioceptive regulation (Sunberg 1973) Central neural: unknown, but may involve increased synaptogenesis, synaptic potential, cortical reorganization for sensorimotor control, especially somatosensory cortex (Kieber et al 2013)
	Maladaptive (Pathophysiological)	Neuromuscular: insufficient motor recruitment, leading to reduced muscle strength (Boucher et al 2006) Metabolic: unknown, but may involve depletion of bioenergetic substrates, metabolic byproduct accumulation, and extracellular fluid accumulation leading to intramuscular edema and erythema Cover: epithelial cell death; collagen and elastin fiber separation from basement membrane (Gray & Titze, 1988); alterations in fluid distribution (Zhang et al 2008); tissue fatigue (Titze 1999; Hunter & Titze 2009) Afferent: higher levels of myotatic reflexive responses resulting in higher tonic tension in laryngeal muscle (Rudolph 1961); reflexive hypersensitivity & hyperresponsiveness Central neural: unknown	Neuromuscular: unknown, but may involve chronic decreases in motor unit recruitment (longer recovery times and continued endurance decrements) or compensatory increases in motor recruitment from surrounding vocal muscles (eg, ventricular compression) (Dabirmoghaddam et al, 2021; Rubin et al, 2014) Metabolic: increased reliance on anaerobic energy substrates and greater oxygen deficits (Nanjundeswaran et al, 2017) Cover: phonotraumatic lesions (Hunter & Titze 2009) Afferent: unknown, but may involve muscle III/IV afferents Central neural: unknown

CONCLUSION AND FUTURE DIRECTIONS

It is important to consider the etiology of any clinical entity to know how best to treat patients—including those with chronic or recurrent vocal fatigue. Unfortunately, our poor understanding of etiological mechanisms underlying clinical presentations of vocal fatigue makes it challenging to identify optimal biophysiological targets and ensure we are providing accurate treatment approaches to our patients. This paper highlights specific areas where future studies on biophysiological mechanisms underlying clinical outcomes (eg, acoustics, aerodynamics) and patient perceptions of vocal fatigue (eg, Vocal Fatigue Index) are needed. A better understanding of relationships between these underlying biophysiological mechanisms and their influences on clinical symptoms and voice outcome metrics are also needed. We proposed several potential mechanisms worth exploring in future work; we have also included several ideas for experimental designs. The objective of this exercise is to facilitate scientific investigations and to provide a platform for academic-clinical discussions. We hope this is the first step towards improving vocal fatigue management, shifting clinical paradigms from symptom-based to etiology-focused approaches.