Vocal Biomechanics

The preponderance of voice problems that are amenable to voice therapy involve some form of abnormality in vocal fold adduction. Phonotraumatic lesions such as nodules, polyps, and nonspecific inflammation consequent on voice use are traceable to hyperadduction resulting from vocal fold impact stress. Adduction causes monotonic increases in impact stress (Jiang and Titze, 1994). In turn, impact stress appears to be a primary cause of phonotrauma (Titze, 1994). Thus, therapy targeting a reduction in adduction is indicated in cases of hyperadduction. Another large group of diagnostic conditions involves hypoadduction of the vocal folds. Examples include vocal fold paralysis, paresis, atrophy, bowing, and nonadducted hyperfunction (muscle tension dysphonia; for a discussion, see Hillman et al., 1989). Treatment that increases vocal fold closure is indicated in such cases.

Voice therapy addresses adductory deviations using a variety of biomechanical solutions. The traditional approach to hyperadduction and its sequelae has targeted the use of widely separated vocal folds and small-amplitude oscillations during voice production; examples are use of a “quiet, breathy voice” (Casper et al., 1989; Casper, 1993) or quiet “yawn-sigh” phonation (Boone and McFarlane, 1993). This general approach is sensible for the reduction of hyperadduction and thus phonotraumatic changes, in that vocal fold impact stress, and phonotrauma, should be reduced by it. There is evidence that the quiet, breathy voice approach is effective in reducing signs and symptoms of phonotraumatic lesions for individuals who use it outside the clinic (Verdolini-Marston et al., 1995). However, individuals may also restrict their use of a quiet, breathy voice extraclinically because it is functionally limiting (Verdolini-Marston et al., 1995).

The traditional approach to hypoadduction has involved “pushing” and “pulling” exercises, which should reduce the glottal gap (e.g., Boone and McFarlane, 1994). Indeed, some data corroborate clinicians' impressions that this approach can increase voice intensity in individuals with glottal incompetence (Yamaguchi et al., 1993).

A more recent approach to treating adductory abnormalities has focused on the use of a single “ideal” vocal fold configuration as the target for both hyperadduction and hypoadduction. The configuration involves barely separated vocal folds, which is “ideal” because it optimizes the trade-off between voice output strength (relatively strong) and vocal fold impact stress, and thus reduces the potential for phonotraumatic injury (Berry et al., 2001). Voice produced with this intermediate laryngeal configuration has been called “resonant voice,” perceptually corresponding to anterior oral vibratory sensations during “easy” voicing (Verdolini et al., 1998). Programmatic approaches to resonant voice training have shown reductions in phonatory effort, voice quality, and laryngeal appearance (Verdolini_-Marston et al., 1995), as well as reductions in functional disruptions due to voice problems in individuals with conditions known or presumed to be related to hyperadduction, such as nodules. Moreover, there is evidence that individuals use this type of voicing outside the clinic more than the traditional “quiet, breathy voice” because it is functionally tractable (Verdolini-Marston et al., 1995). Resonant voice training may also be useful in improving vocal and functional status in individuals with hypoadducted dysphonia. Recent theoretical modeling has indicated that nonlinear source (vocal fold)_–filter (vocal tract) interactions are critical in maximizing voice output germane to resonant voice and other voice types (Titze, 2002).

A relatively small number of clinical cases involve vocal fold elongation abnormalities as the salient feature of the vocal condition. Often, the medical condition involves cricothyroid paresis, although thyroarytenoid paresis may also be implicated. Voice therapy has been less successful in treating such conditions. Other elongation abnormalities are functional, as in mutational falsetto. The clinical consensus is that voice therapy generally is useful in treating mutational falsetto.

Finally, in addition to addressing laryngeal kinematics, voice therapy usually also addresses nonphonatory aspects of biomechanics that influence the vocal fold mucosa. Such issues are addressed in voice hygiene programs (see voice hygiene). Mucosal performance and mucosal vulnerability to trauma are the key concerns. The primary issues targeted are hydration and behavioral control of laryngopharyngeal reflux. Dehydration increases the pulmonary effort required for phonation, whereas hydration decreases it and also decreases laryngeal phonotrauma (e.g., Titze, 1988; Verdolini, Titze, and Fennell, 1994; Solomon and DiMattia, 2000). Thus, hydration regimens are appropriate for individuals with voice problems and dehydration (Verdolini-Marston, Sandage, and Titze, 1994). There is increasing support for the view that laryngopharyngeal reflux plays a role in a wide range of laryngeal diseases, including inflammatory and even neurogenic and malignant disease. Voice therapy can play a supportive role to the medical or surgical treatment of laryngopharyngeal reflux by educating patients regarding behavioral issues such as diet and sleeping position. Some data are consistent with the view that control of laryngopharyngeal reflux can improve both laryngeal appearance and voice symptoms in individuals with a diagnosis of laryngopharyngeal reflux (Shaw et al., 1996; Hamdan et al., 2001). However, vocal hygiene programs alone in voice therapy apparently produce little benefit if they are not coupled with voice production work.

Specific Therapy Approach

Recently, interest has emerged in cognitive mechanisms involved in skill acquisition and factors affecting patient compliance as related to voice training and therapy models. Speech_-language pathologists may train individuals to acquire the basic biomechanical changes described in preceding paragraphs, and others. The traditional approach is eclectic and entails implementing a series of facilitating techniques such as the “yawn-sigh” and “push-pull” techniques, as well as other maneuvers, such as altering the tongue position, changing the loudness of the voice, using chant talk, and using digital manipulation. Facilitating techniques are used by many clinicians and are generally considered effective. However, formal efficacy data are lacking for most of the techniques. An exception is digital manipulation, specifically manual circumlaryngeal therapy (laryngeal massage), used for idiopathic, presumably hyperfunctional dysphonia. Brief courses of aggressive laryngeal massage by skilled practitioners have dramatically improved voice in individuals with this condition (Roy et al., 1997). Also, variants of “yawn-sigh” phonation, such as falsetto and breathy voicing, may temporarily improve symptoms of adductory spasmodic dysphonia and increase the duration of the effectiveness of botulinum toxin injections (Murry and Woodson, 1995).

Several programmatic approaches to voice therapy have been developed, some of which have been submitted to formal clinical studies. An example is the Lee Silverman Voice Treatment (LSVT). This treatment uses “loud” voice to treat not only hypoadduction and hypophonia, but also prosodic and articulatory deficiencies in individuals with Parkinson's disease. LSVT utilizes a predetermined hierarchy of speech tasks in 16 therapy sessions delivered over 4 weeks. In comparison with control and alternative treatment groups, LSVT has increased vocal loudness and voice inflection for as long as 2 years following therapy termination (Ramig, Sapir, Fox et al., 2001; Ramig, Sapir, Countryman et al., 2001). Critical aspects of LSVT that may contribute to its success include a large number of repetitions of the target “loud voice” in a variety of physical contexts.

Another programmatic approach to voice therapy, the Lessac-Madsen Resonant Voice Therapy (LMRVT), was developed for individuals with either hyper_-or hypoadducted voice problems associated with nodules, polyps, nonspecific phonotraumatic changes, paralysis, paresis, atrophy, bowing, and sulcus vocalis. LMRVT targets the use of barely touching or barely separated vocal folds for phonation, a configuration considered to be ideal because it maximizes the ratio of voice output intensity to vocal fold impact intensity (Berry et al., 2001). In LMRVT, eight structured therapy sessions typically are delivered over 8 weeks. Training emphasizes sensory processing and the extension of “resonant voice” to a variety of communicative and emotional environments. Data on preliminary versions of LMRVT indicate that it is as useful as quiet, breathy voice training for sorority women with phonotrauma or the use of amplification for teachers with voice problems in reducing various combinations of phonatory effort, voice quality, laryngeal appearance, and functional status (Verdolini-Marston et al., 1995).

Another programmatic approach to voice therapy for both hyper_-and hypoadducted conditions is called Vocal Function Exercises (VFE; Stemple et al., 1994). This approach targets similar vocal fold biomechanics as LMRVT, that is, vocal folds that are barely touching or barely separated, for phonation. Training consists of repeating maximally sustained vowels and pitch glides twice daily over a period of 4–6 weeks. Carryover exercises to conversational speech may also be used. A 6_-week program of VFE in teachers with voice problems resulted in greater self-perceived voice improvement, greater phonatory ease, and better voice clarity than that achieved with vocal hygiene treatment alone (Roy et al., 2001).

Another program, Accent Therapy, addresses the ideal laryngeal configuration—barely touching or barely separated vocal folds—in individuals with hyper_-and hypoadducted conditions (Smith and Thyme, 1976). Training entails the use of specified rhythmic, prosodically stressed vocal repetitions, beginning with sustained consonants and progressing to phrases and extended speech. The Accent Method is more widely used in Europe and Asia than in the United States.

Electromyographic biofeedback has been reported to be effective in reducing laryngeal hyperfunction and laryngeal appearance in individuals with voice problems linked to hyperadduction (nodules). Also, visual feedback using videoendoscopy may be useful in treating numerous voice conditions; specific clinical observations have been reported relative to ventricular phonation (Bastian, 1987).

Finally, some clinicians have found that sensory differentiation exercises may help in the treatment of repetitive strain injury—one of the fastest growing occupational injuries. Repetitive strain injury involves decreased use of manual digits or voice and pain subsequent to overuse. Attention to sensory differentiation in the treatment of repetitive strain injury is motivated by reports of fused representation for groups of movements in sensory cortex following extensive digit use (e.g., Byl, Merzenich, and Jenkins, 1996).