The larynx, colloquially known as the voice box, is an organ consisting of several cartilages and muscles that connect the upper respiratory tract (nasal cavity, oral cavity, pharynx) to the lower respiratory tract (trachea, bronchi, lungs; Figure 1). The vocal folds (VF) are structures composed of multiple layers (Table 1 for abbreviations used in the article). The epithelium and superficial lamina propria are together referred to as the mucosa or cover. The intermediate and deep layers of the lamina propria are composed mostly of elastin and collagen, respectively. These layers form the vocal ligament. Beneath the ligament is the thyroarytenoid muscle. The ligament and muscle are referred to as the body of the vocal fold. Anteriorly the vocal ligament inserts into the thyroid cartilage via the anterior commissure tendon. Posteriorly the intermediate layer of the lamina propria attaches to the vocal process of the arytenoid via a tendon-like structure (Hirano et al., 1987; Hirano, 1993; Paulsen and Tillmann, 1997). The vocal folds abduct and adduct to act as a sphincter to the lower airway. This allows the larynx to participate in physiologic functions such as promoting airflow during respiration, providing airway protection during deglutition, and facilitating phonation (McHanwell, 2008).

VF motion is controlled by the intrinsic laryngeal muscles (ILM). In humans, these muscles are the thyroarytenoid muscle (TA), lateral cricoarytenoid, posterior cricoarytenoid, interarytenoid, and cricothyroid muscles (Figure 1). Other species demonstrate variations on this. For instance, rodents lack the interarytenoid muscle and instead have the cricoalar and the superior cricoarytenoid muscles (Hernández-Morato et al., 2013a). The ILM contributes to gross movements of the VF, including opening with inhalation and exhalation to allow passage of air between the upper and lower airways, and closing during swallowing to protect the airway from food and saliva. The ILM also contributes to finer-tuned, precise VF movements. This includes micromovements needed to place the VF in the proper position and tension to allow air passing through the larynx to vibrate the vocal folds at specific frequencies. VF position changes constantly during phonation, to facilitate the production of different sounds and pitches (Sasaki, 2006; Song et al., 2009).

The larynx is innervated primarily by two nerves, the superior (SLN) and the recurrent laryngeal nerve (RLN; Figure 1). Both contribute to motor innervation: the external branch of the SLN to the cricothyroid and the RLN to the remaining ILM. These nerves’ contributions to laryngeal sensory innervation appear to vary by species. In humans, the glottic and supraglottic sensation is carried by the internal branch of the SLN while infraglottic sensation is carried by the RLN (McHanwell, 2008). By contrast, in rats, only the SLN carries sensory information (Andrew, 1956; Furusawa et al., 1996; Pascual-Font et al., 2011; Weissbrod et al., 2011; Hernández-Morato et al., 2013a,b).

Adapted to the tasks of the VF, the ILM is among the fastest contracting muscles yet require near instantaneous, constant adjustment (Baken and Orlikoff, 2000; Hoh, 2005; Sasaki, 2006). Such action in turn requires precisely calibrated sensorimotor function. Adjustments of these muscles constantly occur in response to changes in breathing, the sensation of a foreign element such as food, and the need of voice production. These adjustments are influenced by mechano-, chemo-, and baroreceptors of the larynx and airway, as well as aural feedback. While it has been known for some time that somatosensory dysfunction contributes to dysphagia, there is also substantial evidence that disordered laryngeal sensorimotor function plays a role in dysphonia due to VF pathologies, including VF paralysis and spasmodic dysphonia (Sant’Ambrogio et al., 1985; Ludlow et al., 1995; Shiba et al., 1995, 1997; Barkmeier et al., 2000; Jürgens, 2002; Jafari et al., 2003; Ali et al., 2006; Larson et al., 2008; Rosales and Dressler, 2010; Simonyan and Ludlow, 2010, 2012; Hammer and Krueger, 2014; Pitman, 2014; Konczak et al., 2015).

In contrast to the well-studied proprioceptive mechanisms in the body and extremities, proprioception in the larynx is poorly understood despite being studied since the 19th century (Hall, 1833; Head, 1889a,b; Hines, 1927). The following is a literature review detailing the current understanding of the peripheral components of laryngeal proprioception.

The mechanics of proprioception in skeletal muscles are well-described. Two canonical proprioceptive organs, muscle spindles (MuSp) and Golgi tendon organs (GTO), provide sensory input. MuSp are embedded in the muscle body while GTO are found in muscle tendons. MuSp and GTO collect information about muscle length and tension. This information is relayed to the central nervous system by afferent Group Ia, Ib, and II sensory nerve fibers. Subsequently, the efferent arm is activated with alpha motor neurons activating muscle movement while gamma motor neurons set the gain of the MuSp (Gardner and Johnson, 2014).

Joint capsules and ligaments also contain several sensory organs with afferent nerve terminals that carry sensory information including pain, touch, pressure, and stretch. The main encapsulated mechanoreceptors related to proprioception in these organs are Pacini and Ruffini receptors. While Pacini receptors are activated by compression, Ruffini receptors are stimulated by stretch. In contrast, non-encapsulated free nerve endings provide pain-related sensory input from the joints (Gardner and Johnson, 2014).

The presence of canonical intramuscular proprioceptors (MuSp and GTOs) in the ILM is unclear. In particular, despite exploration since the 1960s, evidence of laryngeal MuSp remains controversial (Table 2).

As an alternative to MuSp and GTO, could the ILM have a unique intramuscular proprioceptive apparatus altogether? Brandon et al. (2003a) speculate the TA may contain “spindle-like” mechanoreceptors which convey proprioceptive information and that perhaps modified extrafusal fibers serve proprioceptive purposes. The intrafusal fibers that comprise MuSp are, after all, specialized muscle fibers (Gardner and Johnson, 2014). Additionally, Rossi describes a unique entity consisting of spiral nerve endings around single muscle fibers in the human ILM (Rossi and Cortesina, 1965).

There are other muscles innervated by cranial nerves which also appear to lack canonical proprioceptive organs and in which alternatives have been proposed:

(a)

Extraocular muscles: MuSp are present in the extraocular muscles of some mammals (e.g., humans and sheep) yet absent in others (e.g., some species of monkeys and apes, dogs, cats, and rabbits; Cooper and Daniel, 1949; Lukas et al., 1994; Ruskell, 1999; Blumer et al., 2006; Lienbacher and Horn, 2012) Human extraocular MuSp, though present, are noted to be simplistic in structure (Lienbacher and Horn, 2012). In species without MuSp, terminal nerve ending formations called palisade endings have been proposed as alternative proprioceptors (Billig et al., 1997; Eberhorn et al., 2005; Büttner-Ennever et al., 2006; Lienbacher and Horn, 2012). First described in cats and monkeys in the 1900s, palisade endings are myelinated nerve fibers that enter the muscle centrally, run to the distal tip of the muscle, branch, and then end in the myotendinous junction, surrounding this area in a palisading pattern and terminating on myotendinous cylinders, loose connective tissue caps encapsulating single muscle fiber-tendon junctions (Ruskell, 1999; Lukas et al., 2000; Konakci et al., 2005). However, studies debate whether palisade endings are truly proprioceptive or motor in function. Specifically, studies have shown that some palisade endings nerve terminals contain choline acetyltransferase vesicles and degenerate after injury to extraocular motor nuclei, suggesting an effector role (Lukas et al., 1994; Eberhorn et al., 2005; Büttner-Ennever et al., 2006; Lienbacher and Horn, 2012).

The human larynx contains three paired encapsulated joints: the cricothyroid, cricoarytenoid, and arytenocorniculate joints (McHanwell, 2008). Additionally, the thyroepiglottic ligament joins the epiglottis to the thyroid cartilage. Given these joints’ role in the articular reflexes of the larynx, studies from the 1950s and 60s have investigated the presence of proprioceptors in the laryngeal joints using electrophysiological and histochemical techniques. They demonstrated that laryngeal joints and ligaments contain Pacini and Ruffini organs, and numerous free nerve endings (Andrew, 1954, 1956; Kirchner and Wyke, 1965; McHanwell, 2008; Table 2). Further, in contrast to other joints in the body, Pacinian organs are more abundant than Ruffini organs in the encapsulated laryngeal joints (Kirchner and Wyke, 1965).

Studies investigating the source of sensory innervation to the laryngeal joints show the encapsulated joints receive input from the internal branch of the SLN, and the cricothyroid joint may also receive input from the RLN (Kirchner and Wyke, 1965; Storey, 1968). The thyroepiglottic ligament is also innervated by the internal branch of the SLN (Andrew and Oliver, 1951; Andrew, 1954, 1956).

Many afferent receptors are present in the laryngeal mucosa (Adzaku and Wyke, 1979; Bradley, 2000; Foote and Thibeault, 2021). Electrophysiological studies and staining with mechanoreceptor marker Piezo2 demonstrate that a subset of these receptors is mechanosensitive (Sampson and Eyzaguirre, 1964; Storey, 1968; Davis and Nail, 1987; Widdicombe, 2001; Prescott et al., 2020). While these mucosal mechanoreceptors have not been directly implicated in proprioception, their anatomic distribution and contribution to laryngeal reflexes suggest a potential role in the afferent arm of the laryngeal proprioceptive arc (Table 2).

Mechanoreceptors are not uniformly distributed throughout the laryngeal mucosa and are focused in and around the vibratory and mobile elements of the larynx. Electrophysiological studies in cats and rabbits demonstrate that mechanoreceptors are concentrated in the laryngeal mucosa in the glottal area above and below the vocal folds. Some are sensitive to vibration from the VF and others to changes in pressure from airflow (Sampson and Eyzaguirre, 1964; Storey, 1968; Sant’Ambrogio et al., 1983, 1991; Mathew et al., 1984; Davis and Nail, 1987; Mortola and Piazza, 1987; Shiba et al., 1997; Bradley, 2000; Widdicombe, 2001). Additionally, Piezo2 staining of rodent larynges shows that the aryepiglottic fold and the vocal folds contain clusters of mechanoreceptors (Prescott et al., 2020). Further, conglomerates of VGLUT-positive staining are observed in the mucosa overlying and laryngeal surface of the epiglottis of rats (Soda and Yamamoto, 2012; Takahashi et al., 2016).

Authors also find evidence these mucosal receptors play a role in the laryngeal adductor response. Specifically, Andreatta et al. show that mucosal mechanoreceptors play an important role in the laryngeal adductor response, which is diminished when the larynx is stripped of mucosa (Mårtensson, 1964; Andreatta et al., 2002).

The study of laryngeal proprioception is fragmented in method and species. Nevertheless, a few unifying conclusions can be drawn from the current knowledge. First, canonical proprioceptive organs are rarely expressed in the ILM, aside from the interarytenoid muscle. Second, this is consistent with the finding that other cranial muscles, such as extraocular and facial mimetic muscles, lack MuSp but demonstrate evidence of non-canonical proprioceptive organs. Third, given the lack of canonical elements in ILM, an alternative proprioceptive mechanism may govern the coordination of laryngeal movement, and this requires further study.

IH-M and VY: conceptualization, formal analysis, and writing—original draft. IH-M, VY, and MP: writing—review and editing. IH-M: visualization. MP: supervision, project administration, and funding acquisition. All authors contributed to the article and approved the submitted version.