| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
NEUROSCIENCE |
1 Laryngeal and Speech, Section, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
(Received 18 August 2007;
accepted after revision 16 October 2007;
first published online 25 October 2007)
Corresponding author C. L. Ludlow: Laryngeal and Speech Section, National Institute of Neurological Disorders and Stroke, Bldg. 10 Room 5D 38, 10 Center Drive MSC 1416, Bethesda, MD 20892-1416, USA. Email: ludlowc{at}ninds.nih.gov
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Laryngeal adductor responses typically can be elicited by stimulation of the afferents in the glottis such as mechanoreceptors in the mucosa overlying the arytenoid cartilages both in the cat and in humans (Andreatta et al. 2002; Bhabu et al. 2003). When a servomotor is used to displace the arytenoid cartilage by 325 µm for 10 ms on one side, which will stretch the laryngeal mucosa and the thyroarytenoid muscle, an ipsilateral R1 response is elicited in the cat (Andreatta et al. 2002). When the same stimulus was applied after the mucosa was peeled from the glottis on one side, the response was abolished indicating that mechanoreceptors in the mucosa may be involved in eliciting this response (Andreatta et al. 2002). In the human, single air puffs to the mucosa overlying the arytenoid cartilages will produce a brief vocal fold closure (Aviv et al. 1999) as a result of a bilateral thyroarytenoid muscle response beginning around 80–100 ms (Bhabu et al. 2003). Because both the R2 response to electrical stimulation of the superior laryngeal nerve (SLN) and the response to air puffs to the mucosa are bilateral with similar latencies (Bhabu et al. 2003; Ludlow, 2005), responses to mechanoreceptor stimulation in the mucosa probably involve the same pathways as the R2 response to electrical stimulation of the SLN afferents. In addition, when interstimulus intervals between a conditioning stimulus and a test stimulus are less than 1 s, very similar conditioning effects are seen in R2 responses to electrical stimuli to the SLN (Ludlow et al. 1995) and in thyroarytenoid muscle responses to conditioned air puff stimuli to the laryngeal mucosa (Kearney et al. 2005). On the other hand, the R1 response is more frequently elicited by electrical stimulation of SLN afferents and is relatively consistent compared to the R2 response. During conditioning protocols the amplitude of the conditioned R1 response is unchanged while the R2 response is reduced (Ludlow et al. 1995).
Tasks involving changes in vocal fold position produce different types of movements. During phonation for speech, voice control requires rapid opening and closing of the vocal folds for voice onset and offset. Here the larynx is under cortical control (Schulz et al. 2005) and the laryngeal muscles must be precisely controlled (Poletto et al. 2004). These muscles contain few, if any, spindles (Brandon et al. 2003) and no stretch reflexes can be elicited from the intrinsic laryngeal muscles (Loucks et al. 2005). Vocal fold movements during speech probably affect mechanoreceptors in the mucosa which, when stimulated at rest, will produce reflex closure (Bhabu et al. 2003).
Laryngeal muscle control during forced inhalation requires hyper-abduction of the vocal folds to allow high flow rates through the glottis into the trachea. On the other hand, during effort closure (valsalva) the vocal folds are hyper-adducted to produce a tight seal for maintaining chest wall pressure during heavy lifting or defecation. If increased sensory input interacts with motor control demands, thyroarytenoid muscle responses may be affected differently dependent upon such tasks. To test this hypothesis, we compared the frequency of occurrence and amplitude of R1 and R2 responses evoked by SLN stimulation during phonation, humming, forced inhalation, effort closure, in comparison with rest. Because voice, humming and effort closure require increased thyroarytenoid muscle firing, the increased drive to motor neurons during such tasks could facilitate the LAR. We tested three alternative hypotheses regarding the modulation of the LAR during task production (Fig. 1 ). First, if there is no suppression of the LAR, response amplitude during a task should equal the sum of baseline muscle activity during the task added to the reflex response at rest (Fig. 1B). Second, if the cortical control interferes with the reflex response, then the LAR evoked by stimulation of the SLN during the task may be reduced from the LAR evoked at rest possibly by the increased baseline activity occurring during the task performance masking the LAR (Fig. 1C). Third, if the LAR is unaltered during task performance from rest then it may momentarily suppress motor neuron firing during task performance and equal the LAR at rest (Fig. 1D).
|
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Selection criteria for healthy volunteers included: normal laryngeal structure and function when examined by an otolaryngologist; no current or past history of speech, voice, neurological, or psychiatric disorders; and no medications that could alter central nervous system function prior to or during the study.
Procedures
Informed written consent to participate in a National Institutes of Health Internal Review Board-approved research protocol was obtained from each participant before study. All of the experiments conformed with the Declaration of Helsinki. After an earth electrode was attached, participants were supine with the neck extended and a small amount of 2% Xylocaine with adrenaline (epinephrine) (1: 10 000) was injected subcutaneously over the region of the hyoid bone and the upper margin of the thyroid lamina. To reduce discomfort during EMG needle insertion, Xylocaine was injected subcutaneously over the cricothyroid membrane.
Methods were similar to previous experiments (Ludlow et al. 1995). Bipolar electromyography (EMG) needle electrodes were used to locate the laryngeal muscles using standard techniques (Hirano & Ohala, 1969) before placing bipolar hooked wires contained in 27-gauge 37.5 mm needles into the thyroarytenoid (TA) muscles bilaterally. Verification gestures were used to assure that the electrode was in the targeted muscle. For example for the TA muscles, verification gestures consisted of increased muscle activity during phonation of a vowel (/i/) and during effort closure. Subjects were trained to produce effort closure at the level of the larynx by inhaling with their mouth open, holding the air at the level of the vocal folds and making a short vocal sound during the release to assure that closure was at the level of the vocal folds. When only onset and offset bursts occurred during phonation indicative of lateral cricoarytenoid muscle placement, the electrode was moved to a more anterior and superior position to be in the TA muscle. Once the correct location was achieved, the needles were removed and the hooked wires remained in place throughout the study.
Peak to peak 2 V saw tooth calibration signals were recorded on each EMG channel. The two EMG channels were band-pass filtered between 100 and 5000 Hz and recorded along with the stimulus trigger and the subject's voice on a TEAC multiple-channel FM recorder with VHS tape at 15 in s–1. After electrode placement, the lighting was dimmed and noise minimized to allow the participant to relax for approximately 5 min prior to recording muscle activity during quiet respiration. Next electrodes were inserted on one side in the region where the internal SLN penetrates through the thyrohyoid membrane (Ludlow et al. 1992). Hooked wire electrodes were inserted proximal to either side of the SLN. The rostral anode electrode and the caudal cathode electrode were usually between 0.5 and 1.5 cm apart and 1–3.5 cm in depth, depending on a subject's neck dimensions. A third hooked wire electrode served as an earth to reduce the stimulation artifact in the EMG recordings, and was inserted between the stimulation and EMG electrodes close to the midline of the superior border of the thyroid cartilage. A DISA 15E07 current stimulator was used to deliver a 1 ms rectangular current pulse in milliamps (mA). Because we have previously shown that when current intensity to the SLN is systematically increased, R1 response amplitudes increase in a linear fashion (Yamashita et al. 1997), we used the supramaximal current levels for each subject to ensure that response amplitude would not vary due to changes in stimulus characteristics throughout the experiment. The current was increased until the R1 response in the ipsilateral TA muscle no longer increased, that is, had reached supramaximal levels, between 3 and 6 mA. In some instances, the highest levels approved by the internal review board were used (7 mA) if the supramaximal level was not reached.
Six trials were presented for each type of task and a total of five tasks were studied: quiet inspiration (rest), phonation, humming, forced inhalation, and effort closure. Task order was randomized across subjects. Only one stimulus (a single 1 ms electrical pulse) was presented per task performance. Between trials, the participants relaxed for at least 20 s. During each trial, the participant was instructed to begin the task and a single SLN stimulus was presented at random manually by the experimenter between 1 and 5 s later to prevent subject anticipation of the timing of the stimulus. Participants were instructed to ignore the effects of SLN stimulation and to continue performing the task during and after the stimulation. Stimuli were presented only on inspiration during the rest condition to control for response variation due to changes in the breathing cycle. The responses at rest were checked at the beginning and end of the experiment to ensure that neither the stimulation nor EMG electrodes had moved over the course of the study.
Signal analysis
The trigger, EMG, and calibration signals were digitized off-line at 5 kHz after using anti-aliasing, low-pass filtering at 2 kHz. Linear interpolation of the calibration signals was used to convert the signals into microvolts (µV). The signals were corrected for DC offset by averaging non-rectified data over 20 s of quiet respiration. Next, to correct for electrode impedance differences, the signals were rectified and the minimum activation level (usually < 1 µV) between motor unit firings during quiet respiration was measured and subtracted from the EMG signal for a channel. The mean rectified TA muscle activity in the absence of stimulation (baseline mean) was computed from a 20 ms window immediately prior to each stimulation (Fig. 2 ).
|
Automated signal-processing algorithms used the onset and offset times marked for each R1 and R2 response to measure: (1) latency (2) duration, and (3) integral (area under the curve). To normalize variances, the mean baseline and integral measures were natural log transformed.
Mean response integral and frequency (percentage of stimuli producing a response) were measured from the six trials for a task for the ipsilateral R1 and the ipsilateral R2 responses. To test whether volitional tasks reduced response probability, we computed the difference in percentage occurrence between the response frequency at rest and during a task (Rest – Task) separately for the R1 and R2 responses.
Statistical analyses
For each task compared to rest, change in response frequency and change in latency were compared to zero using single group t tests with SYSTAT 11 (Systat Software Inc.). A Bonferroni-corrected P value of 
0.025 was used because the same hypothesis was being tested during each task on two measures: latency and frequency.
We then compared the difference in the mean response integral using three hypothetical models of response modulation by task adjusting for changes in baseline tone. Because three specific hypotheses were tested regarding response modulation during tasks compared to rest, paired within subject t tests were used to test each of the specific hypotheses. Further, because we were not able to complete all of the tasks on all of the participants, an ANOVA across all tasks could not be used. Instead we conducted within-subject paired t tests to test each of the three hypotheses in contrast with rest for each task independently.
The first hypothesis was whether the response during a task was equal to the response at rest added to the task activity. Here the amount of task activity that would increase a response integral was equal to the mean baseline amplitude (µV) for the task multiplied by the duration of the response (ms) at rest added to the response integral at rest. If this hypothesized amount equalled the response integral during a task then the difference between the two would be zero. On the other hand, if a negative difference score was obtained that would indicate that the response integral during the task was less that the response at rest added to task baseline activity (Fig. 1B).
To test for responses being masked by increased baseline activity during a task, the response integral at rest minus the mean baseline amplitude during a task multiplied by the duration of the rest response, was subtracted from the response integral during the task (Fig. 1C). If the difference was zero then masking had occurred but if a positive difference occurred then the response integral during the task was not masked by the baseline activity (Fig. 1C).
Finally, to test whether task activity was blocked during the response, the integral of the muscle response at rest was subtracted from the integral of the response during the task (Fig. 1D). If the difference between the two was close to zero then task activity was blocked during the response during task performance.
For each hypothesis on each task, the difference between the hypothetical response and the measured response was compared to zero using a within-subject paired t test with a Bonferroni corrected P value of 0.013 because the same hypothesis was tested on each of four tasks.
Regression analyses
We performed a stepwise regression to determine if the response integral during a task (the dependent variable) could be predicted by (1) the response integral at rest, (2) the duration of the response at rest, or (3) the change in the task baseline from rest baseline. To normalize the response integrals across subjects, we computed the percentage response increase during a task as follows:
%Response increase during the task = ((log of the response integral during the task – log of the mean baseline for the task)/(log of the mean baseline for the task)) x 100.
In a similar fashion, the predictor variables were also normalized across subjects. The mean response at rest was calculated as the difference between the log of the response integral at rest minus the log of the mean baseline at rest divided by the log of the mean baseline at rest and multiplied by 100. Similarly, the other normalized predictor variable was the percentage increase in the mean baseline during the task minus the mean baseline at rest divided by the mean baseline at rest multiplied by 100. Separate regression analyses were performed for ipsilateral R1 and ipsilateral R2 responses using all of the tasks combined with a significance level of P 0.025 for each response type, R1 and R2.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Task effects on R1 and R2 frequency of occurrence
Stimulation of the superior laryngeal nerve (SLN) during the rest condition produced an average R1 latency of 16.76 ± 1.07 ms (± S.E.M.) and an R2 latency of 68.35 ± 2.68 ms. R2 response frequency at rest was 66.7% on the side ipsilateral to stimulation and less consistent than R1 which occurred in 100% of the rest trials. R2 responses on the side contralateral to stimulation only occurred in 40.5% of trials at rest. R1 response latency was similar during the different tasks and rest (P > 0.05) while R1 response frequency showed no group effect on any of the tasks (P > 0.05) but decreased in frequency relative to rest in four of the 10 subjects on effort closure and in some subjects during forced inhalation and humming. R1 response frequency did not decrease, however, during phonation relative to rest (Fig. 3A ). R2 response frequency relative to rest could only be contrasted on the ipsilateral side as there were too few responses on the contralateral side. Ipsilateral R2 frequency of occurrence was reduced from rest during the effort closure (t = –2.70, P = 0.018) and humming (t = –2.64, P = 0.039) but not during forced inhalation (t = –2.11, P = 0.051) or phonation (t = –1.66, P = 0.074) (Fig. 3B). Because R2 response frequency was reduced during most tasks, changes in R2 amplitude and latency could not be tested for statistical significance across tasks.
|
As expected, the mean baseline level (µV) of TA muscle activity during some of the tasks increased relative to the mean baseline at rest (Fig. 4
). This was statistically significant (P 0.013) on t tests for effort closure (t = 5.833, P = 0.001) and phonation (t = 5.381, P = 0.003) but not during humming (t = 3.274, P = 0.031) or forced inhalation (t = 1.169, P = 0.295).
|
To test for a summation of R1 response with muscle activity during a task, we compared the R1 response area under the curve (integral) during a task with the R1 response at baseline added to the mean baseline during the task multiplied by the response duration at rest (ms). All of the mean values were negative (Table 1 ) with P < 0.0005 demonstrating a significant reduction from the expected value if the R1 response at rest had summated with the increased muscle activity for a task.
|
Finally, we compared the R1 response integral during the task with the R1 response integral at rest. The mean values were close to zero with P values all greater than 0.013 for each task (Table 1). These results demonstrated that the response integrals for the R1 were unchanged by task (Fig. 5 ). The same computations could not be performed for the ipsilateral or contralateral R2 responses because the degrees of freedom were too low for statistical testing during forced inhalation and humming.
|
The multiple regression analyses for predicting the normalized response amplitude during a task (measured as the percentage increase over task baseline) examined three normalized predictors, the percentage change in the mean response integral at rest from the mean baseline at rest, the response duration at rest and the percentage change in the task baseline from rest baseline for the R1 responses. The R1 analysis had a significant overall result with multiple R = 0.595, F = 7.397, P = 0.003 and included two predictive factors; a positive relationship with the response at rest (t = 3.028, P = 0.005) and an inverse relationship with the percentage change in the baseline of the task from rest baseline (t = –3.736, P = 0.001) (Fig. 6 ).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The R1 seemed to transiently disrupt the increased baseline motor neuron firing during task performance similarly for each task; the R1 responses that occurred during each task were not changed in amplitude compared to rest (Fig. 5). Although during forced inhalation only one subject had increased thyroarytenoid muscle activity compared to rest (Fig. 4), the R1 response integral remained the same between rest and task regardless of task type. The inverse relationship between the normalized response amplitude and percentage increase from baseline for a task, indicated that the response during a task was greater when the baseline was less. This occurred during forced inhalation (Fig. 6B), which also related to the response amplitude at rest (Fig. 6A). The lack of change in R1 response integrals regardless of task-related differences in thyroarytenoid muscle firing (Fig. 4), suggests that the task-related increases in motor neuron firing were momentarily blocked during the LAR responses.
Because of previous reports of presynaptic inhibition of cutaneous afferent input during active limb movement (Seki et al. 2003), the laryngeal afferent responses could have been expected to be suppressed during volitional voice and respiratory tasks. During voluntary limb movement, presynaptic inhibition of afferent inputs to the spinal circuits involving motor neurons in the lumbar spinal cord was shown in monkeys (Seki et al. 2003). The suppression occurred prior to and during both extension and flexion movement indicating that they were unaffected by the direction of movement.
Stimulation of the SLN during voluntary tasks produced a decrease in R2 response frequency in two tasks: humming and effort closure. Therefore, some interference may have occurred in the R2 pathway which was not apparent in the R1 pathway because no significant reduction occurred in R1 frequency during task performance. The R2 decrease on some tasks did relate to the differences in vocal fold function across the tasks studied. For example, during effort closure the thyroarytenoid muscles are maximally contracted for hyperadduction, and only partially closed during humming. The LAR would only minimally change vocal fold position during either of these tasks when the vocal folds are already closed. During forced inhalation the vocal folds are maximally open and the LAR would oppose the task opening of the vocal folds, while during phonation there is vocal fold adduction similar to humming. Because humming and phonation had similar levels of thyroarytenoid muscle contraction but different effects on R2 occurrence, the changes in R2 frequency did not seem to relate to thyroarytenoid muscle activity for task performance.
The interaction between movement and sensory inputs may differ across systems with a variety of input effects on movement. During jaw closing, trigeminal inputs that are not painful produce a suppression of activity in the masseter, which increases with stimulus intensity in duration and degree (Komiyama et al. 2006). With tendon vibration of an antagonist muscle (i.e. the masseter) during jaw opening, both speech and non-speech movements are reduced in amplitude (Ostry et al. 1997; Loucks & De Nil, 2001). In addition, stretch reflexes occur at some jaw opening positions but decrease at others (Wang et al. 2007). Peripheral inputs therefore may affect jaw muscle control depending upon their movement extent and direction.
In the LAR, however, the R1 was not suppressed in this study even when subjects were instructed to continue to perform the task and ignore the SLN stimulation effects. The ipsilateral R1 involves a relatively direct oligoneuronal brainstem reflex pathway (Ambalavanar et al. 2004). In cats, this pathway begins with the projection of primary afferents onto neurons in the interstitial subnucleus of the nucleus tractus solitarii (NTS) with projection via interneurons in the lateral tegmental field to motor neurons for the laryngeal muscles in the nucleus ambiguus (Ambalavanar et al. 2004). Second order afferents in the NTS are modulated during rapid inputs to the NTS (Sessle, 1973) but this was avoided here by having a 20 s delay between stimulus presentation. Any changes in R1 responses in the TA muscle must occur within this pathway. Because the R1 responses were unchanged, it is unlikely that suppression of incoming afferents to the R1 pathway was responsible for the reduced frequency of R2 responses during some tasks in this study.
The lack of R1 modulation during voice and respiratory tasks found here differed markedly from our previous results using electrical SLN stimulation to study the LAR before, during and after a swallow, when both R1 and R2 response amplitudes were reduced during swallowing (Barkmeier et al. 2000). Barkmeier et al. demonstrated that R2 response frequency was significantly reduced and R1 responses showed a similar reduction. The swallowing central pattern generator in the medulla can be triggered by intense or frequent SLN stimulation (Sang & Goyal, 2001) but may not depend upon further afferent input for completion of the pharyngeal pattern of swallowing (Jean, 1984). Therefore, the effects of further SLN inputs during a swallow may be limited once swallowing is initiated. A temporary bilateral block of SLN input significantly interferes with the motor execution of swallowing in humans (Jafari et al. 2003), indicating that volitional control may be interfered with by a peripheral loss of sensory input. Suppression of the LAR during swallowing may have occurred either due to a presynaptic suppression of sensory input at the medulla after swallowing initiation or a descending cortico-bulbar suppression affecting the LAR pathway during volitional swallowing. Neither of these mechanisms appeared to be operating during the voice and respiratory tasks studied here.
The role of continuous afferent feedback during voice production is not well known. Movement-induced afferent feedback may be less important during voice production than swallowing because audition has an important role in voice control (Sapir et al. 1983; Burnett et al. 1997, 1998). In a previous study we were unable to elicit a TA muscle response to muscle stretch during phonation or at rest (Loucks et al. 2005) suggesting that muscle spindle stretch does not have a role in regulating laryngeal muscle control. Reduction in R2 frequency was significant only during effort closure and humming which both involve contact or vibration between the vocal folds that could excite the mechanoreceptors in the mucosa during execution. The other two tasks, phonation and forced inhalation, however, also involve afferent feedback, due to vocal fold vibration and airflow, respectively. There was no indication therefore that differences in afferent feedback between tasks played a role in decreasing R2 response frequency.
Previous studies have demonstrated inhibition of the R2 response, but not the R1, within the LAR pathway with paired stimulation conditioning; conditioned R2 responses are reduced in amplitude and frequency of occurrence relative to unconditioned responses (Ludlow et al. 1995; Kearney et al. 2005). We have hypothesized that conditioning effects with repeated and rapid stimulation could be the mechanism responsible for the lack of LAR occurrences during voice production for speech (Kearney et al. 2005). On the other hand, mucosal stimulation by air pressure may not normally be intense enough to elicit R1 responses while electrical stimulation of afferents is more intense and probably involves additional afferents responsible for R1 responses. This may explain why R1 responses, which are not modified by conditioning effects, are not evident during voice or respiratory tasks in response to vocal fold vibration, closure, air pressure or flow.
One possible explanation for the lack of modification of the LAR during this study could be that electrical stimulation of the SLN was used rather than more natural stimulation such as air puff stimulation of the mucosa. During task performance, access to the laryngeal mucosa for air puff stimulation via nasoendoscopy is limited and unreliable, and for that reason we used electrical stimulation to examine whether or not there was modulation of the LAR as we had done previously during swallowing (Barkmeier et al. 2000). Studies of afferent presynaptic inhibition in limb movement also used electrical stimulation of the afferent nerves (Seki et al. 2003). Therefore, it was reasonable to expect that our results would be comparable to studies of afferent modulation during volitional limb movements in primates.
The reduced occurrence of the R2 response during effort closure and humming may involve mechanisms that are present in the R2 pathway and not in the R1 response pathway. The SLN afferents project to both solitary tract neurons and reticular neurons in the cat (Sessle, 1973). With presentation of repeated sensory stimuli, Sessle found postsynaptic inhibition of NTS interneurons demonstrating the presence of inhibitory interneurons in this circuit (Sessle, 1973). He also found that neurons in the reticular circuit had longer latencies than those in the solitary tract and were subject to more widespread inhibitory influences.
An inverse relationship was found between the R1 amplitude and the percentage increase in muscle activity over baseline. This suggests that the amount of motor neuron firing for the task may have played some role in reducing the R1 amplitude.
However, motor neuron facilitation during volitional task-related performance should lead to an increase in the reflex response amplitude and a decrease in the latency of the response (Capaday & Stein, 1987). The lack of a significant increase in the LAR R1 response amplitude during volitional tasks suggests that some gating decreased the volitional drive without decreasing the activity of the motor neurons in response to SLN input. Furthermore, because response latency did not change, there was no shift in motor neuron threshold. In fact, because the LAR was unchanged from rest, there was some support for the conclusion that the LAR transiently gated the increased motor neuron activity due to task performance. This differs from the report during limb movement in primates that found there was no inhibition of the primate corticospinal pathway by peripheral input (Jackson et al. 2006). These differences between the two systems attest to the stability of this robust protective reflex during volitional tasks.
In conclusion, this study of the LAR during the performance of voluntary voice and respiratory tasks showed that this protective mechanism for the upper airway was highly resistant to suppression. This demonstrates the predominance of upper airway protection during voluntary use of the upper airway for voice, forced inhalation or chest wall stabilization. These results also suggest that the interactions in the brain stem between cortico-bulbar descending voluntary control and afferent reflexes for upper airway protection differ from spinal afferent modulation during voluntary limb movements.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Andreatta RD, Mann EA, Poletto CJ & Ludlow CL (2002). Mucosal afferents mediate laryngeal adductor responses in the cat. J Appl Physiol 93, 1622–1629.
Aviv JE, Martin JH, Kim T, Sacco RL, Thomson JE, Diamond B & Close LG (1999). Laryngopharyngeal sensory discrimination testing and the laryngeal adductor reflex. Ann Otol Rhinol Laryngol 108, 725–730.[Medline]
Barkmeier JM, Bielamowicz S, Takeda N & Ludlow CL (2000). Modulation of laryngeal responses to superior laryngeal nerve stimulation by volitional swallowing in awake humans. J Neurophysiol 83, 1264–1272.
Bhabu P, Poletto C, Mann E, Bielamowicz S & Ludlow CL (2003). Thyroarytenoid muscle responses to air pressure stimulation of the laryngeal mucosa in humans. Ann Otol Rhinol Laryngol 112, 834–840.[Medline]
Brandon CA, Rosen C, Georgelis G, Horton MJ, Mooney MP & Sciote JJ (2003). Staining of human thyroarytenoid muscle with myosin antibodies reveals some unique extrafusal fibers, but no muscle spindles. J Voice 17, 245–254.[CrossRef][Medline]
Burnett TA, Freedland MB, Larson CR & Hain TC (1998). Voice F0 responses to manipulations in pitch feedback. J Acoust Soc Am 103, 3153–3161.[CrossRef][Medline]
Burnett TA, Senner JE & Larson CR (1997). Voice F0 responses to pitch-shifted auditory feedback: a preliminary study. J Voice 11, 202–211.[CrossRef][Medline]
Capaday C & Stein RB (1987). A method for simulating the reflex output of a motoneuron pool. J Neurosci Meth 21, 91–104.[CrossRef][Medline]
Deleyiannis FW, Gillespie M, Bielamowicz S, Yamashita T & Ludlow CL (1999). Laryngeal long latency response conditioning in abductor spasmodic dysphonia. Ann Otol Rhinol Laryngol 108, 612–619.[Medline]
Hirano M & Ohala J (1969). Use of hooked-wire electrodes for electromyography of the intrinsic laryngeal muscles. J Speech Hear Res 12, 362–373.[Medline]
Jackson A, Baker SN & Fetz EE (2006). Tests for presynaptic modulation of corticospinal terminals from peripheral afferents and pyramidal tract in the macaque. J Physiol 573, 107–120.
Jafari S, Prince RA, Kim DY & Paydarfar D (2003). Sensory regulation of swallowing and airway protection: a role for the internal superior laryngeal nerve in humans. J Physiol 550, 287–304.
Jean A (1984). Control of the central swallowing program by inputs from the peripheral receptors. A review. J Auton Nerv Syst 10, 225–233.[CrossRef][Medline]
Kearney PR, Poletto CJ, Mann EA & Ludlow CL (2005). Suppression of thyroarytenoid muscle responses during repeated air pressure stimulation of the laryngeal mucosa in awake humans. Ann Otol Rhinol Laryngol 114, 264–270.[Medline]
Komiyama O, Wang K, Svensson P, Arendt-Nielsen L & De Laat A (2006). Reproducibility of the masseteric exteroceptive suppression period using stimulus-response curves. J Oral Rehabil 33, 741–748.[CrossRef][Medline]
Loucks TMJ & De Nil LF (2001). The effects of masseter tendon vibration on nonspeech oral movements and vowel gestures. J Speech Lang Hear Res 44, 306–316.
Loucks TM, Poletto CJ, Saxon KG & Ludlow CL (2005). Laryngeal muscle responses to mechanical displacement of the thyroid cartilage in humans. J Appl Physiol 99, 922–930.
Ludlow CL (2005). Central nervous system control of the laryngeal muscles in humans. Respir Physiol Neurobiol 147, 205–222.[CrossRef][Medline]
Ludlow CL, Schulz GM, Yamashita T & Deleyiannis FW (1995). Abnormalities in long latency responses to superior laryngeal nerve stimulation in adductor spasmodic dysphonia. Ann Otol Rhinol Laryngol 104, 928–935.[Medline]
Ludlow CL, Van Pelt F & Koda J (1992). Characteristics of late responses to superior laryngeal nerve stimulation in humans. Ann Otol Rhinol Laryngol 101, 127–134.[Medline]
Odom JL (1993). Airway emergencies in the post anesthesia care unit. Nurs Clin North Am 28, 483–491.[Medline]
Ostry DJ, Gribble PL, Levin MF & Feldman AG (1997). Phasic and tonic stretch reflexes in muscles with few muscle spindles: human jaw-opener muscles. Exp Brain Res 116, 299–308.[CrossRef][Medline]
Poletto CJ, Verdun LP, Strominger R & Ludlow CL (2004). Correspondence between laryngeal vocal fold movement and muscle activity during speech and nonspeech gestures. J Appl Physiol 97, 858–866.
Sang Q & Goyal RK (2001). Swallowing reflex and brain stem neurons activated by superior laryngeal nerve stimulation in the mouse. Am J Physiol Gastrointest Liver Physiol 280, G191–G200.
Sapir S, McClean MD & Larson CR (1983). Human laryngeal responses to auditory stimulation. J Acoust Soc Am 73, 315–321.[CrossRef][Medline]
Sasaki CT & Suzuki M (1976). Laryngeal reflexes in cat, dog and man. Arch Otolaryngol 102, 400–402.
Schulz GM, Varga M, Jeffires K, Ludlow CL & Braun AR (2005). Functional neuroanatomy of human vocalization: an H215O PET study. Cereb Cortex 15, 1835–1847.
Seki K, Perlmutter SI & Fetz EE (2003). Sensory input to primate spinal cord is presynaptically inhibited during voluntary movement. Nat Neurosci 6, 1309–1316.[CrossRef][Medline]
Sessle BJ (1973). Excitatory and inhibitory inputs to single neurones in the solitary tract nucleus and adjacent reticular formation. Brain Res 53, 319–331.[CrossRef][Medline]
Wang K, Lobbezoo F, Svensson P & Arendt-Nielsen L (2007). Influence of jaw gape on EMG of jaw muscles and jaw-stretch reflexes. Arch Oral Biol 52, 562–570.[CrossRef][Medline]
Widdicombe JG (1995). Neurophysiology of the cough reflex. Eur Respir J 8, 1193–1202.[Abstract]
Yamashita T, Nash EA, Tanaka Y & Ludlow CL (1997). Effects of stimulus intensity on laryngeal long latency responses in awake humans. Otolaryngol Head Neck Surg 117, 521–529.[CrossRef][Medline]
|
Acknowledgements |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
