Respiratory and laryngeal contributions to maximum phonation duration.

Maximum phonation duration (MPD) is a common assessment procedure in speech-language pathology. However, the specific contributions of the respiratory and phonatory components of the speech-production mechanism to this task are not typically assessed. Six women and 6 men with normal speech and voice were monitored for lung volume during a standard MPD task, and for laryngeal airway resistance (Rlaw) during a modified MPD (slow syllable-repetition) task. On average, subjects used 90% of their vital capacity (VC) for their best MPD trial. There was no systematic relation between MPD and VC for these subjects. Rlaw was strongly correlated with MPD for the men (r(s) = 0.886 for /a/; r(s) = 0.829 for /i/), but not for the women. Rlaw increased linearly as lung volume decreased (slope > 0.15) for a subset of trials (32%). This was a common pattern for four of the subjects. The clinical utility of MPD to assess breathing for speech is questioned because of the lack of association between MPD and VC, and some atypical laryngeal-valving strategies.

Effects of a vocally fatiguing task and systemic hydration on phonation threshold pressure.

Phonation threshold pressure (PTP), effort for speaking, and vibratory closure pattern were assessed in 4 women with normal untrained voices after 2 hours of loud reading. PTP generally increased after this vocally fatiguing task at conversational pitch and 10%, 50%, and especially 80% of the pitch range. Increased systemic hydration by drinking water appeared to attenuate and/or delay the elevation of PTP for 3 subjects, at least at the highest pitch tested. Effort for speaking increased consistently throughout the loud reading task and subsequently decreased after 15 minutes of vocal silence. Upon videostroboscopic examination of the larynx, 3 subjects demonstrated spindle-shaped vibratory closure patterns on occasion after loud reading. The results provide preliminary support for increasing water consumption to reduce or delay some vocal-function changes after prolonged loud phonation in untrained speakers.

Strength, endurance, and stability of the tongue and hand in Parkinson disease.

Weakness and fatigue in the orofacial system often are presumed to contribute to the dysarthria associated with neuromotor disorders, although previous research findings are equivocal. In this study, tongue strength, endurance, and stability during a sustained submaximal effort were assessed in 16 persons with mild to severe Parkinson disease (PD) and a perceptible speech disorder. The same measures were taken from one hand for comparison. Only tongue endurance was found to be significantly lower in these participants than in neurologically normal control participants matched for sex, age, weight, and height. Analyses of data from a larger sample comprising the present and retrospective data revealed lower-than-normal tongue strength and endurance in participants with PD. No significant correlations were found between tongue strength and endurance, interpause speech rate, articulatory precision, and overall speech defectiveness for the present and previously studied participants with PD, bringing into question the influence of modest degrees of tongue weakness and fatigue on perceptible speech deficits.

Changes in normal speech after fatiguing the tongue.

Detrimental effects of tongue fatigue on speech have been assumed to exist based on neuromotor speech disorders. However, to address whether fatigue is a contributing cause to impaired speech requires an experimental protocol with an uncomplicated population. This study induced tongue fatigue in eight neurologically normal persons and examined changes in speech perceptually and acoustically. The fatigue task consisted of repeated cycles of 6 s of sustained maximum voluntary contraction and 4 s of rest until 50% of maximum strength could not be achieved for three consecutive cycles. Participants then produced speech that was weighted heavily with lingual-palatal consonants. Perceptual analyses of the speech revealed a statistically significant deleterious effect of induced tongue fatigue on speech precision and an incomplete reversal of this effect after a recovery period. Acoustically, the first and third spectral moments (mean and skewness) of the spectral energy for /see text/, /see text/, and /see text/ differed significantly after fatigue but in directions opposite to a priori predictions. Tendencies were found for decreased stop-closure duration and increased voice onset time for /see text/ after fatigue. Supplemental analyses revealed decreased second formant (F2) frequency for /see text/ and /see text/ and flattened F2 transition for the diphthong /see text/ after fatigue. These results indicate disruption of tongue positioning and transitioning for lingual-palatal consonants during speech after prolonged strenuous tongue exercises.

Midbrain regions for eliciting vocalization by electrical stimulation in anesthetized dogs.

Eliciting vocalization in anesthetized dogs by midbrain stimulation is a useful procedure for studies of laryngeal and respiratory physiology. The goal of this report has been to construct a canine stereotaxic "map" that would allow investigators to locate midbrain stimulus sites producing vocalization. Motor responses to electrical stimulation at currents of 1.5 mA or less were observed at 1,158 stimulus sites throughout the midbrains of 8 dogs. Vocalization was observed at 213 stimulus sites. The highest probability of observing vocalization was for sites located 6 to 10 mm anterior, 6 to 7 mm lateral, and 5 to 8 mm dorsal to earbar zero. The vocalization region most likely consists of axons arising in the midbrain periaqueductal gray and coursing through the adjacent tegmentum; low-threshold sites are close to the medial lemniscus. The relationship between stimulus sites at which vocalization was elicited and sites producing other motor responses is described.

Comparison of concentric needle versus hooked-wire electrodes in the canine larynx.

BACKGROUND: The use of a specific electrode type in laryngeal electromyography has not been standardized. Laryngeal electromyography is usually performed with hooked-wire electrodes or concentric needle electrodes. Hooked-wire electrodes have the advantage of allowing laryngeal movement with ease and comfort, whereas the concentric needle electrodes have benefits from a technical aspect and may be advanced, withdrawn, or redirected during attempts to appropriately place the electrode. OBJECTIVES: This study examines whether hooked-wire electrodes permit more stable recordings than standard concentric needle electrodes at rest and after large-scale movements of the larynx and surrounding structures. A histologic comparison of tissue injury resulting from placement and removal of the two electrode types is also made by evaluation of the vocal folds. METHODS: Electrodes were percutaneously placed into the thyroarytenoid muscles of 10 adult canines. Amplitude of electromyographic activity was measured and compared during vagal stimulation before and after large-scale laryngeal movements. Signal consistency over time was examined. Animals were killed and vocal fold injury was graded and compared histologically. RESULTS: Waveform morphology did not consistently differ between electrode types. The variability of electromyographic amplitude was greater for the hooked-wire electrode (p < 0.05), whereas the mean amplitude measures before and after large-scale laryngeal movements did not differ (p > 0.05). Inflammatory responses and hematoma formation were also similar. CONCLUSIONS: Waveform morphology of electromyographic signals registered from both electrode types show similar complex action potentials. There is no difference between the hooked-wire electrode and the concentric needle electrode in terms of electrode stability or vocal fold injury in the thyroarytenoid muscle after large-scale laryngeal movements.

Sense of effort and the effects of fatigue in the tongue and hand.

Fatigue and increased effort are common symptoms for people with movement disorders and dysarthria, but they are rarely quantified. In an attempt to develop a clinically useful and physiologically meaningful measure of fatigue, we used a task that involves sustaining a target effort level without visual feedback while squeezing a bulb connected to a pressure transducer. In the first experiment, 12 healthy young adults performed the constant-effort task with the tongue and the preferred hand at 3 submaximal levels of effort. The resulting pressure declined over time as a negative exponential function with a nonzero asymptote. In the second experiment, 6 subjects performed the constant-effort task before and after acutely fatiguing the tongue and hand. The rate of pressure decline was significantly greater after fatigue. One possible mechanism for the characteristic negative exponential function is that it reflects a constant descending drive from higher centers in the CNS to the appropriate motoneuron pools. Thus, this technique may elucidate the contribution of central fatigue to normal and disordered speech.

Fundamental frequency and tracheal pressure during three types of vocalizations elicited from anesthetized dogs.

Electrical stimulation of the midbrain was used to elicit a variety of vocalizations from six anesthetized dogs. This study was conducted to investigate the ranges of and relationships between fundamental frequency of the vocalizations (F0) and tracheal pressure (Pt) produced during the vocalizations. The vocalizations were described according to type (growl, howl, and whine); F0 and Pt, as well as patterns of laryngeal muscle activity, were examined for each vocalization type. Natural-sounding growl and howl vocalizations were elicited from five dogs; three dogs also produced whines. With few exceptions, F0 was categorically different for the three vocalization types (low for growls, average for howls, very high for whines). Pt values overlapped for the three vocalization types, although, on average, howls were produced with greater Pt than growls. Patterns and degrees of laryngeal muscle activity varied across and within vocalization types, but general findings were consistent with the presumed function of most of the muscles. Laryngeal muscle activity may help explain some of the variability in the acoustic and aerodynamic data.

Effect of subglottic pressure on fundamental frequency of the canine larynx with active muscle tensions.

The relation between subglottic pressure and the fundamental frequency of vocal fold vibration was studied by means of evoked phonation in an in vivo canine model. The evoked-phonation model involved electrical stimulation of the midbrain that resulted in consistent responses by respiratory and laryngeal musculature, accompanied by phonation. The dynamic stiffness properties of the vocal folds, especially the "cover," were investigated by delivering various amounts of air pressure to the larynx from an opening in the trachea. The fundamental frequency of vocal fold vibration increased linearly with subglottic pressure. The slopes ranged from 22.4 to 118.7 Hz per kilopascal in 7 animals. The results indicated that the dependence of fundamental frequency on subglottic pressure is a passive mechanical phenomenon.

Interference between normal vibrato and artificial stimulation of laryngeal muscles at near-vibrato rates.

A stabilized tremor hypothesis for vocal vibrato is investigated. The stabilizer is assumed to be a mechanical oscillator that may contain reflex loops. Artificial stimulation of the cricothyroid muscle in one subject showed a well-defined resonance curve of this peripheral oscillator at approximately 5.0 Hz. Combined artificial stimulation with natural vibrato showed that the vibrato could be entrained by a peripheral stimulus, provided the two frequencies are separated by no more than approximately +/- 0.5 Hz. This suggests that vibrato frequencies are not "hard-wired" centrally, even though a collection of centrally generated tremors may serve as excitation to the peripheral oscillator.

Modulation of fundamental frequency by laryngeal muscles during vibrato.

The variations in voice fundamental frequency (F0) that occur during vibrato production may be produced, at least in part, by modulation of laryngeal muscle activity. We have quantified this relation by using a cross-correlation analysis of the changes in F0 during vibrato and the changes either in motor unit firing rate or in gross electromyographic activity from the cricothyroid (CT) and the thyroarytenoid (TA) muscles. Two trained amateur tenors provided the data. Correlations were generally quite strong (mean r for the CT was 0.72 for singer 1 and 0.50 for singer 2; mean r for the TA was 0.31 for singer 2), thus providing support for previous evidence that fundamental frequency modulation in vibrato involves active modulation of the laryngeal motoneuron pool, especially by the CT muscle. In addition, phase delays between muscle modulation and changes in fundamental frequency were substantial (averaging approximately 130 degrees for the CT and 140 degrees for the TA). This finding may help provide insight regarding the mechanisms responsible for the production of vibrato.

Effect of lung volume on voice onset time (VOT).

This investigation was designed to test the hypothesis that voice onset time (VOT) varies as a function of lung volume. Recordings were made of five men as they repeated a phrase containing stressed /pi/ syllables, beginning at total lung capacity and ending at residual volume. VOT was found to be longer at high lung volumes and shorter at low lung volumes in most cases. This finding points out the need to take lung volume into account when using VOT as an index of laryngeal behavior in both healthy individuals and those with speech disorders.

Speech breathing in Parkinson's disease.

Breathing was investigated in 14 male subjects with Parkinson's disease and 14 healthy male control subjects. Kinematic, spirometric, acoustic, and pressure data were used to assess function during resting tidal breathing, reading aloud, and monologue production. Data were collected at two times during the drug cycle for subjects with Parkinson's disease. During resting tidal breathing, subjects with Parkinson's disease, on average, had a faster breathing rate, greater minute ventilation, and smaller relative contribution of the rib cage to lung volume change than did healthy control subjects. During speech breathing, rib cage volume was smaller and abdominal volume was larger at initiation of the breath groups for subjects with Parkinson's disease than for healthy control subjects. Subjects with Parkinson's disease produced fewer words and spent less time producing speech per breath group and tended to have a faster interpause speech rate than did healthy control subjects. There was no difference between groups for duration of inspirations between speech breath groups. Oral pressure was lower for subjects with Parkinson's disease but estimated tracheal pressure did not differ between the two subject groups. Few differences were found between the two times in the drug cycle for resting and speech breathing. Results provide indirect evidence for reduced relative compliance of the rib cage to the abdomen for subjects with Parkinson's disease as compared to healthy control subjects. In addition, the results support the possibility of inadequate valving of the air stream for subjects with Parkinson's disease.

Respiratory and laryngeal function during whispering.

Established procedures for making chest wall kinematic observations (Hoit & Hixon, 1987) and pressure-flow observations (Smitheran & Hixon, 1981) were used to study respiratory and laryngeal function during whispering and speaking in 10 healthy young adults. Results indicate that whispering involves generally lower lung volumes, lower tracheal pressures, higher translaryngeal flows, lower laryngeal airway resistances, and fewer syllables per breath group when compared to speaking. The use of lower lung volumes during whispering than speaking may reflect a means of achieving different tracheal pressure targets. Reductions in the number of syllables produced per breath group may be an adjustment to the high rate of air expenditure accompanying whispering compared to speaking. Performance of the normal subjects studied in this investigation does not resemble that of individuals with speech and voice disorders characterized by low resistive loads.

Recording single motor unit activity of human nasal muscles with surface electrodes: applications for respiration and speech.

A method is presented for recording nasal single motor unit (SMU) potentials from the skin surface using a 3-pole 'branched' bipolar electrode. Stable, high-quality recordings of single motor unit activity were obtained for up to 3 h. Branched electrode arrays were capable of locating an SMU's maximal voltage point within 5 mm. We examined nasal SMU discharge patterns in relation to respiration in 9 adult humans. The majority of SMUs which discharged during quiet breathing began firing late in expiration and ceased firing in mid-inspiration, other SMUs discharged only during expiration, and a few fired continually with frequency modulation during breath cycles. With increased ventilation, new SMUs were recruited, and previously active SMUs increased the frequency and duration of their discharge. We examined the discharge of 13 units (5 adults) which discharged during speech but were never active during quiet or moderately increased breathing. Some of these SMUs fired during production of nasal consonants, and others were active for articulations involving facial movements (bilabial stops, labio-dental fricatives, and vowels produced with lip movement). By providing information about motor neuron recruitment which cannot be obtained from gross EMG recordings, surface recording of unit potentials may be useful in studying the central nervous control of the nasal upper airway, face, and neck for respiration and speech.

Laryngeal configuration and constriction during two types of whispering.

Laryngeal functioning during the production of whispered speech is not well understood. The efficacy of whispering as a means of voice rest is a common clinical concern that is unresolved. This investigation examined vocal-fold configuration, glottal size, and airway constriction by supraglottal structures during whispering. Ten normally speaking adults produced consonant-vowel syllables with three different vowels while whispering in low-effort and high-effort manners. The larynx was visualized through fiberoptic endonasolaryngoscopy, and the views were recorded on videotape. Analysis and descriptions of the data revealed that low-effort and high-effort whispering were differentiated to a small extent by vocal-fold adjustments and to a somewhat larger degree by supraglottal constriction. However, for each dependent variable, individual subject differences tended to be considerably larger than any systematic effects due to whisper type or vowel.