Microstimulation in Different Parts of the Periaqueductal Gray Generates Different Types of Vocalizations in the Cat

Open AccessPublished:March 06, 2020DOI:https://doi.org/10.1016/j.jvoice.2020.01.022

      Summary

      In the cat four different types of vocalization, mews, howls, cries, and hisses were generated by microstimulation in different parts of the periaqueductal gray (PAG). While mews imply positive vocal expressions, howls, hisses, and cries represent negative vocal expressions. In the intermediate PAG, mews were generated in the lateral column, howls, and hisses in the ventrolateral column. Cries were generated in two other regions, the lateral column of the rostral PAG and the ventrolateral column of the caudal PAG. In order to define the specific motor patterns of the mews, howls, and cries, the following muscles were recorded during these vocalizations; larynx (cricothyroid, thyroarytenoid, and posterior cricoarytenoid), tongue (genioglossus), jaw (digastric), and respiration muscles (diaphragm, internal intercostal, external, and internal abdominal oblique). During these mews, howls, and cries we analyzed the frequency, intensity, activation cascades power density, turns, and amplitude analysis of the electromyograms (EMGs). It appeared that each type of vocalization consists of a specific circumscribed motor coordination. The nucleus retroambiguus (NRA) in the caudal medulla is known to serve as the final premotor interneuronal output system for vocalization. Although neurochemical microstimulation in the NRA itself also generated vocalizations, they only consisted of guttural sounds, the EMGs of which involved only small parts of the EMGs of the mews, howls, and cries generated by neurochemical stimulation in the PAG. These results demonstrate that positive and negative vocalizations are generated in different parts of the PAG. These parts have access to different groups of premotoneurons in the NRA, that, in turn, have access to different groups of motoneurons in the brainstem and spinal cord, resulting in different vocalizations. The findings would serve a valuable model for diagnostic assessment of voice disorders in humans.

      Key Words

      INTRODUCTION

      Vocalizations in cats are regarded as expressions of their emotional state.
      • John ER
      • Chesler P
      • Bartlett F
      • Victor I
      Observation learning in cats.
      • Bradshaw JWS
      The Behavior of the Domestic Cat.
      • Landsberg G
      Feline behavior and welfare.
      • Bradshaw J
      • Cameron-Beaumont C
      The signaling repertoire of the domestic cat and its undomesticated relatives.
      • Beaver BV
      Feline Behavior: A Guide for Veterinarians.
      • Yeon SC
      • Kim YK
      • Park SJ
      Differences between vocalization evoked by social stimuli in feral cats and house cats.
      Cats produce different vocalizations, such as mews, howls, growls, and cries.
      • John ER
      • Chesler P
      • Bartlett F
      • Victor I
      Observation learning in cats.
      • Bradshaw JWS
      The Behavior of the Domestic Cat.
      • Landsberg G
      Feline behavior and welfare.
      Mewing is an expression of a positive emotion, because cats convey it during states of playfulness, joy, food-seeking, attention-seeking, and mating. In contrast, growl is considered to be an expression of rage, while crying and howling are expressions of negative emotions, connoting pain, fear, threat, anxiety, and states of health problems.
      • John ER
      • Chesler P
      • Bartlett F
      • Victor I
      Observation learning in cats.
      • Bradshaw JWS
      The Behavior of the Domestic Cat.
      • Landsberg G
      Feline behavior and welfare.
      ,
      • Yeon SC
      • Kim YK
      • Park SJ
      Differences between vocalization evoked by social stimuli in feral cats and house cats.
      • Nicastro N
      • Owren MJ
      Classification of domestic cats (felis catus) vocalizations by naïve and experienced human listeners.
      • Darwin CR
      The expression of the emotions in man and animals.
      Neurochemical microstimulation studies have shown that these vocalizations can be elicited in the midbrain periaqueductal gray (PAG).
      • Bandler R
      • Carrive P
      Integrated defence reaction elicited by excitatory amino acid injection in the midbrain periaqueductal gray region of the unrestrained cat.
      • Carrive P
      • Dampney RA
      • Bandler R
      Somatic and autonomic integration in the midbrain of the unanesthetized decerebrate cat: a distinctive pattern evoked by excitation of neurones in the subtentorial portion of the midbrain periaqueductal grey.
      • Zhang SP
      • Davis PJ
      • Bandler R
      Brain stem integration of vocalization: role of midbrain periaqueductal gray.
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      • Subramanian HH
      • Holstege G
      The midbrain periaqueductal gray changes the eupneic respiratory rhythm into a breathing pattern necessary for survival of the individual and of the species.
      Lesions in the PAG in cats can cause mutism
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      • Subramanian HH
      • Holstege G
      The midbrain periaqueductal gray changes the eupneic respiratory rhythm into a breathing pattern necessary for survival of the individual and of the species.
      • Skultety FM
      The behavioral effects of destructive lesions of the periaqueductal gray matter in adult cats.
      • Adametz J
      • O'Leary JL
      Experimental mutism resulting from periaqueductal lesions in cats.
      ; rendering them incapable of expressing vocalizations, although the forebrain structures are intact. These results demonstrate that the PAG is the critical neural structure for vocal expression, not only in cats
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      ,
      • Subramanian HH
      • Holstege G
      The midbrain periaqueductal gray changes the eupneic respiratory rhythm into a breathing pattern necessary for survival of the individual and of the species.
      ,
      • Holstege G
      Anatomical study of final common pathway for vocalization in the cat.
      but also in humans.
      • Alexander MP
      Chronic akinetic mutism after mesencephalic-diencephalic infarction: remediated with dopaminergic medications.
      ,
      • Holstege G
      • Subramanian HH
      Two different motor systems generate human speech.
      The next question is how the different types of vocalizations are encoded in the PAG. Are the neurons that produce positive vocal expressions located in different regions of the PAG than the neurons that produce negative vocal expressions, and do the laryngeal and abdominal motor patterns distinctly differ between positive and negative vocalizations?
      Although in previous studies
      • Zhang SP
      • Davis PJ
      • Bandler R
      Brain stem integration of vocalization: role of midbrain periaqueductal gray.
      ,
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      various muscles have been recorded during vocalizations in the cat, the precise activation patterns of laryngeal, tongue, jaw, and respiratory muscles and their integration during positive and negative vocalizations, are not known. In all likelihood, positive and negative vocal expressions possess different laryngeal, oral and respiratory motor patterns. Investigating PAG-triggered vocal behavior in the cat will open doors to evaluate laryngeal control during human speech and to understand human otolaryngological diseases and avenues of rehabilitation.
      In this study, the PAG was topographically mapped to investigate where in the PAG the mews, cries, growls, and howls are encoded. Subsequently, the recruitment patterns and interrelationships between various laryngeal [posterior cricoarytenoid (PCA), cricothyroid, and thyroarytenoid], tongue (genioglossus), mouth-opening (digastric), and respiratory muscles (internal and external abdominal obliques, internal intercostals, and the crural diaphragm) were analyzed during different vocal expressions.
      Holstege
      • Holstege G
      • Subramanian HH
      Two different motor systems generate human speech.
      and Holstege and Subramanian
      • Holstege G
      • Subramanian HH
      Two different motor systems generate human speech.
      have shown that the PAG has access to the laryngeal motoneurons via the premotor interneurons in the nucleus retroambiguus (NRA), located in the caudal medulla. This PAG-NRA-motoneuronal pathway represents the final common pathway for vocalization. Previous studies
      • Zhang SP
      • Davis PJ
      • Bandler R
      Brain stem integration of vocalization: role of nucleus retroambigualis.
      ,
      • Subramanian HH
      • Holstege G
      The nucleus retroambiguus control of respiration.
      report that sound production can also be generated by stimulation in the NRA. Subramanian and Holstege
      • Subramanian HH
      • Holstege G
      The nucleus retroambiguus control of respiration.
      called these NRA vocalizations “guttural sounds devoid of emotional quotient” because they could not be interpreted as either a mew, howl, or cry. In this study, we examined the integration between the laryngeal, oral, and respiratory muscle activities during PAG- and NRA-induced sound in order to understand how the PAG generates the positive and negative emotional expressions.

      MATERIALS AND METHODS

       Use of Cats for This Study

      Data were collected from the same unanesthetized spontaneously breathing precollicularly decerebrated adult cats (n = 8) as used in previous investigations
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      ,
      • Subramanian HH
      • Holstege G
      The nucleus retroambiguus control of respiration.
      ,
      • Subramanian HH
      • Holstege G
      Midbrain and medullary control of postinspiratory activity of the crural and costal diaphragm in vivo.
      approved by The University of Sydney institutional Animal Care Ethics Committee. These experiments were wholly undertaken in the period 1997–2000 at The University of Sydney. No additional cat work was undertaken elsewhere.

       Surgery

      Cats, weighing between 2.2 and 4.5 kg, were anesthetized in a box filled with a mixture of isoflurane and oxygen. Following the induction, the anesthesia was maintained through a facemask with the animal breathing spontaneously, while femoral arterial and venous catheters were inserted. Next, an endotracheal catheter was inserted, which was used for subsequent continued administration of isoflurane. The cat's head was secured in a stereotaxic frame with the body suspended from the frame by straps. In order to avoid the depressing effect of anesthesia, the cats were decerebrated at the precollicular level. To gain access to the midbrain, two burr holes were drilled in the skull on either side of the sagittal sinus. An occipital craniotomy was performed to allow access to the caudal brainstem. The dura was then incised and the medial part of the cortex was suctioned. Following ligation and removal of a portion of the sagittal sinus, precollicular decerebration was carried out using suction diathermia, a surgical technique that uses pulsations of electrical energy to generate heat and cauterize blood vessels to prevent excessive bleeding while suctioning brain tissue. Lowering the mean arterial pressure from 100–130 mm Hg to 65–70 mm Hg by raising the level of anesthesia reduced bleeding during decerebration. All brain tissue rostral to the superior colliculus including the entire diencephalon was removed. After completion of decerebration, anesthesia was discontinued. The cat started to breath spontaneously and the mean arterial pressure returned to 100–130 mm Hg within 30–60 minutes. A thermostatic infrared lamp was employed to maintain the animal body temperature at 37.5–38.5°C. The end tidal CO2 (measured by means of a Morgan 901 gas analyzer Rowe Scientific Ltd, Sydney, Australia) was intermittently monitored. Fluid supplements were administered via the femoral intravenous catheter. All animals had intact vagal nerves, were spontaneously breathing and not tracheotomized.

       EMG Recording

      Unilateral EMG activity was recorded using bipolar Teflon-coated, multistranded stainless-steel electrodes, bared for 2 mm at their tips. The electrodes were surgically implanted in the laryngeal muscles [cricothyroid, thyroartenoid, PCA (Figure 1)], tongue (genioglossus), jaw (digastric)], and respiration involved muscles (crural diaphragm, internal intercostal, and internal and external abdominal oblique). EMG electrode positions were verified following postmortem examination of the respective muscles. Tracheal pressure was recorded through a 19-gauge needle inserted into the trachea and connected to a differential pressure transducer. Mean arterial pressure was recorded through the catheter placed in the femoral artery and obtained from the low pass filtered pressure pulses. The mouth of the cat was kept open and vocalization was recorded with a microphone placed 10 cm from the mouth.
      FIGURE 1
      FIGURE 1Schematic overview of the various laryngeal muscles in the cat.

       Chemical Stimulation of the PAG and the NRA

      PAG and NRA stimulation sites were selected based upon extensive stereotaxic chemical mapping previously undertaken.
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      ,
      • Subramanian HH
      • Holstege G
      The midbrain periaqueductal gray changes the eupneic respiratory rhythm into a breathing pattern necessary for survival of the individual and of the species.
      ,
      • Subramanian HH
      • Holstege G
      The nucleus retroambiguus control of respiration.
      ,
      • Subramanian HH
      • Holstege G
      Midbrain and medullary control of postinspiratory activity of the crural and costal diaphragm in vivo.
      Excitatory amino acid [D, L-homocysteic acid (DLH), Sigma Australia] microinjections (200 mM, 30–60 nL) were used for chemical stimulation of the PAG and the NRA. Homeostasis parameters of the animals (n = 8) during eupnea are listed in Table 1.
      TABLE 1Homeostasis Parameters of the Animals (n = 8) During Eupnea
      Respiratory rate (breaths/minute)40 ± 4
      Inspiration duration (Ti; seconds)0.60 ± 0.05
      Expiration duration (Te; seconds)0.90 ± 0.05
      Arterial pressure (mm Hg)105 ± 5
      Heart rate (beats/minute)199 ± 5
      Arterial blood pH7.39 ± 0.05
      TABLE 2Comparative Differences in the Activity of the Diaphragm and Laryngeal Muscles During the Mew, Howl, and Cry
      Mew vs CryMew vs HowlHowl vs Cry
      SignificantP valueSignificantP valueSignificantP value
      ThyroarytenoidYes0.00468Yes0.00125No0.47570
      DiaphragmYes0.00100Yes0.00103No0.42097
      CricothyroidYes0.00041Yes0.00032Yes0.00011
      Microinjecting DLH is a standard method for selective stimulation of neuronal cell bodies within the central nervous system.
      • Fries W
      • Zieglgansberger W
      A method to discriminate axonal from cell body activity and to analyze ‘silent’ cells.
      ,
      • Goodchild AK
      • Dampney RAL
      • Bandler R
      A method for evoking physiological responses by stimulation of cell bodies, but not axons of passage, within localized regions of the central nervous system.
      For microinjections of DLH, a single-barrel glass-micropipette (tip diameter 10–30 μm) was inserted into the PAG and into the NRA, guided by stereotaxic coordinates. A pressure system (Picospritzer II, Parker Instrumentation Parker Inc, NH, USA) was used to deliver the microinjections. The volume injected was determined using a precalibrated scale. To test the reproducibility of the vocalization generated in each region of the PAG, in each stereotaxic position the injections were repeated three times (with an interval of 25 minutes between the injections to eliminate any residual effects of the previous injection). The same method was employed for testing the reproducibility of vocalization generated in the NRA. Rhodamine microspheres were added to the DLH solution for later histological identification of the injection sites. In all the animals both the PAG and NRA were chemically stimulated. At the end of each experiment, the animal was deeply anesthetized and perfused transcardially with 0.9% saline followed by 4% paraformaldehyde in phosphate buffer (pH = 7.2). After perfusion the brain was removed and stored in 4% formaldehyde for 2 hours, after which it was transferred to a 30% sucrose/formaldehyde mixture for at least 48 hours in order to prevent ice crystal formation. The midbrain was cut on a freezing microtome into 50 μm coronal sections. The injection sites marked with rhodamine microspheres were identified using fluorescent microscopy and were represented on standard drawings according to Berman's
      • Berman AL
      The Brainstem of the Cat. A Cytoarchitectonic Atlas with Stereotaxic Coordinates.
      cat stereotaxic atlas.

       Data Analysis

      The band-pass filtered (0.1–5 kHz) EMG activity of various muscles, as well as tracheal pressure, mean arterial pressure, and voice. They were recorded on the “Pulse Code Modulator” (PCM); (A.R Vetter and Co CA, USA) and played back for analyses using the PCM/Powerlab/Apple-Macintosh system. The AD Instruments differential bioamplifiers (Bio Amp) were used for measurement of EMG signals. MacLab Chart software (AD Instruments, Sydney, Australia) running on the Apple-Macintosh computer was used for signal analysis. Using MacLab software (AD Instruments, Sydney) ensemble averages were derived from amplified (×1000), low pass filtered signals using sampling rates of 20,000/s. All the muscle activities were rectified and averaged within and across animals. The diaphragm EMG signal was used for measurement of Ti and Te.
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      Statview software was used for computation of respiratory parameters; the duration of inspiration (Ti) and expiration (Te) and of the respiratory frequency (RF). Changes to these respiratory parameters during specific vocal episodes (such as mew, howl, and cry) were averaged. Considering the distinctiveness of each type of vocal behavior, parametric analysis to assess statistical significance in data from single and multiple trigger vocal episodes were undertaken. Statistical comparisons were carried out using analysis of variance (ANOVA). Significant differences between the mean values were detected using Scheffe least difference test and a probability of P < 0.05 was considered significant. Chart records have been transformed into Adobe Illustrator software data representation.

       Power Density and Frequency Spectral Analysis

      The spectrum module of the chart software was used to plot the power density and frequency spectra of various laryngeal, oral, and respiratory muscles during specific vocalizations. The power density spectrum uses a discrete Fast Fourier Transform algorithm (Chart software, AD Instruments, Australia) to convert data from time to frequency domains. We selected a Fast Fourier Transform size of 1 kHz for representation of the spectrum in the frequency domain. All individual power spectral windows were averaged and the spectra (Fast Fourier Transforms) were presented as connected points. The X-axis represents a frequency spread from 0 to 1 kHz while the Y-axis represents arbitrary units.

       Turns and Amplitude Measurements

      This EMG quantification method allows assessment of the number and size of motor unit potentials (MUPs) at various force levels.
      • Berman AL
      The Brainstem of the Cat. A Cytoarchitectonic Atlas with Stereotaxic Coordinates.
      A “turn” occurs at the peak of the MUP signal, with successive turns separated by ≥100 μV (to exclude low-amplitude peaks generated by noise and electrical interference). Quantified turns and mean amplitude per turn were measured for ≥ 10 epochs of cricothyroid and thyroarytenoid muscle during mew, cry, and howl. For this, the EMG of cricothyroid and thyroarytenoid activity during mew, howl, and cry was recorded via a 37-mm concentric needle electrode that was inserted into these muscles. A ground electrode was placed over the skin of the ear. The LabChart system was used to filter the signal between 20 Hz and 10 kHz. Motor unit recruitment tracings were recorded with sweep speeds of 10 ms per division. A gain of 200 μV per division was used. Quantified turns and mean amplitude per turn were measured via off-line analysis of the digitized data.

      RESULTS

       Topographical Map of Vocalizations Generated in the PAG

      DLH stimulation in the PAG produced three types of vocalizations: mews, cries, and howls. These vocalizations were topographically organized within the PAG (Figure 2). Mews were generated in the lateral column and howls in the ventrolateral column of the intermediate PAG. Cries were generated in two other regions, the lateral column of the rostral PAG and the ventrolateral column of the caudal PAG. Hisses, an “unvoiced” form of vocalization, were generated in the same area from which the howls were elicited, but never from the same areas that produced either mews or cries. Only hissing or only growl episodes were never elicited.
      FIGURE 2
      FIGURE 2The PAG stimulation sites and the responses evoked. A. Midbrain PAG schematics based on Berman's atlas,
      • Berman AL
      The Brainstem of the Cat. A Cytoarchitectonic Atlas with Stereotaxic Coordinates.
      dm, dorsomedial column; dl, dorsolateral column; lat, lateral column; vlat, ventrolateral column; Dr, dorsal raphe nucleus. B. Coronal sections of the PAG
      • Berman AL
      The Brainstem of the Cat. A Cytoarchitectonic Atlas with Stereotaxic Coordinates.
      showing rhodamine-stained histologically identified stimulation sites in the PAG that produced the mew, howl, cry, and hiss. The localization is based on multiple injections.

       Duration, Frequency, and Decibel Level of Vocalizations

      The duration of a mew was less than one second, the duration of a howl 1–2 seconds, while the duration of a cry was much longer, 4–6 seconds. The average peak fundamental frequency level of the mews was 115 Hz, of the cries 280 Hz, and of the howls 740 Hz (Figure 3).
      FIGURE 3
      FIGURE 3Illustrates the peak frequency and decibel levels of the mew, howl, and cry. X-axis shows frequency in Hertz (Hz), while Y-axis is represented as arbitrary unit (AU). Sound level is expressed in decibels (Db).

       The Laryngeal, Oral, and Respiratory System During Eupnea

      During the inspiratory phase of eupnea the PCA muscle was rhythmically active (Figure 4). The muscle showed preinspiratory activation, that is, started activity just before the diaphragm onset. It stopped activity time-locked with the end of the inspiratory phase of the diaphragm, that is, just before the postinspiratory activation of the diaphragm. The PCA was not active during the expiratory phase. The other two laryngeal muscles, the cricothyroid and thyroarytenoid, were not active during eupnea. The genioglossus muscle was also not active during eupnea, but following PAG stimulation it showed immediate tonic activation and became phasic during the vocal phase (Figure 5). Eupnea was also devoid of abdominal (Figure 5) and internal intercostal expiratory muscle activation. PAG-induced vocalization generated phasic recruitment of the external (Figure 5) and internal abdominal oblique and the internal intercostal muscles.
      FIGURE 4
      FIGURE 4The activation patterns of the PCA and diaphragm muscles and their power density spectrum during eupnea. Note the preinspiratory activation of the PCA. It is not active during either postinspiration (post-I) or expiration.
      FIGURE 5
      FIGURE 5The activation patterns of the genioglossus, external oblique, and diaphragm muscles during eupnea and following PAG-induced vocalization. Note the tonic activation of the genioglossus during eupnea and its conversion into phasic activation during vocalization. The external oblique muscle is not active during eupnea but is recruited during vocalization.

       Activation Patterns of Laryngeal and Respiratory Muscles During the Mew

      The mew was characterized by the phasic activation of the cricothyroid muscle (Figure 6). Its onset and offset were time-locked with the start and the end of the mew. The thyroarytenoid started 0.3–0.5 seconds before the onset of the mew and ended at the same time as the mew. The PCA muscle did not show any activity during the mew. The genioglossus muscle also started firing 0.2–0.3 seconds before the onset of the mew, showed strong phasic activity during the mew and continued to show low-level tonic firing during the subsequent inspiratory phase. In terms of expiratory muscle activation, the external and internal abdominal oblique muscles and the internal intercostal muscles (Figure 7) as well as the abdominal oblique muscles were activated prior to the onset of the mew. The internal intercostal and internal abdominal oblique muscles ended their activity at the same time as the mew, while the external abdominal oblique muscle continued to be active at a low level during almost 75% of the subsequent inspiratory phase following the mew (Figure 7).
      FIGURE 6
      FIGURE 6Activation patterns of the thyroarytenoid, cricothyroid, and genioglossus muscles during the mew.
      FIGURE 7
      FIGURE 7Activation patterns of the internal intercostal and the internal and external oblique abdominal muscles during the generation of mews.

       Activation Patterns of Laryngeal and Respiratory Muscles During the Howl

      The howls were between 1.5 and 2.5 seconds long. The cricothyroid activity was phasic and time-locked with the duration of the howl. Both the thyroarytenoid and external abdominal oblique started activity 0.2 and 0.5 seconds before the beginning of the howl (Figure 8). After the howl ended the thyroarytenoid continued its activity at a low level, while the external abdominal oblique ended almost at the same time as the howl. The howls were sometimes interspersed with the hiss. During the hiss the thyroarytenoid was fairly activated, but the cricothyroid and external abdominal oblique only to a very limited extent, while the crural diaphragm was not activated (Figure 8). In between two howls without hiss the crural diaphragm sometimes generated three short inspirations instead of one (Figure 9).
      FIGURE 8
      FIGURE 8During the howl, the thyroarytenoid, cricothyroid, and external abdominal oblique were activated but not the crural diaphragm. During the hiss, only the thyroarytenoid was activated, while the crural diaphragm activity was blocked.
      FIGURE 9
      FIGURE 9Between howls without hisses, the crural diaphragm was subdivided into three short activations.

       Activation Patterns of Laryngeal and Respiratory Muscles During the Cry

      Of all the vocalizations, the cry had the longest duration, 4–6 seconds. The thyroarytenoid showed a completely different type of discharge pattern during crying as compared to the mew and howl. It started at least 1 second before the beginning of the cry and exhibited in the beginning a decrementing and later a constant discharge pattern (Figure 10). In contrast, the crural diaphragm showed a very strong augmenting discharge pattern with its peak intensity at the end of the cry. Following the vocal outflow, the diaphragm resumed its normal inspiratory activity (Figure 10). The activity patterns of the cricothyroid, external abdominal oblique, internal abdominal oblique, and internal intercostal muscles were similar to that of the patterning seen during the mew or howl.
      FIGURE 10
      FIGURE 10During the cry, the thyroarytenoid showed a decrementing and the crural diaphragm an incrementing activity.

       Activation Patterns of the Digastric Muscle

      The digastric (mouth opening) muscle did not fire during vocalization, irrespective whether it was a mew, howl, or cry. However in between two vocal phases it showed multiple (2–6) phasic activities (Figure 11). After the end of the vocalization sequence, the digastric muscle was phasically active for several seconds before ending firing.
      FIGURE 11
      FIGURE 11The digastric muscle fired rhythmically between mew vocalizations. Note that it continued its rhythmic activity for over 5 seconds following the cessation of the vocalization episode.

       Muscular Activities During Vocalization Generated in the Nucleus Retroambiguus (NRA)

      Vocalizations could only be generated in the rostral part of the NRA. These stimulations never produced equivalents of the mew, cry, or howl as generated in the PAG, but activated the cricothyroid and the thyroarytenoid muscles. Both muscles started before the start of vocalization and the cricothyroid ended its activity after the vocalization (Figure 12). The genioglossus muscle only showed low-level tonic and no phasic activity during the rostral NRA-generated vocalization. Digastric and abdominal muscle activations were not induced from the rostral NRA.
      FIGURE 12
      FIGURE 12Activation patterns of the thyroarytenoid, cricothyroid, and diaphragm muscles during NRA-induced vocalization. Both thyroarytenoid and cricothyroid muscles were activated and deactivated at the same time. Sound onset occurred after the activation of the thyroarytenoid and cricothyroid muscles.

       Quantification of EMG for the Mew, Howl, and Cry

      The EMG data show how the cascades of activation are represented for the mew, howl, and cry over a “1 second” signal induction period (Figure 13A). During the vocal expiration phase, the expiratory muscles (internal and external abdominal obliques) were the first to be activated, beginning 300–500 ms before the vocalization signal onset (Figure 13B). They were followed by the activation of the thyroarytenoid and genioglossus muscles, 250–300 ms before the vocalization signal onset. The cricothyroid muscle was time-locked with the onset of vocalization. Figure 14 demonstrates the intensity versus frequency for muscle activation during vocalization. The peak frequency of activation of the diaphragm for all three vocals, mew, howl, and cry was between 100 and 300 Hz. The laryngeal muscles exhibited very sharp contours for the three vocals. The thyroarytenoid showed sharpest gradient in the intensity, its frequency spread being 200 and 300 Hz. The thyroarytenoid intensity was highest during the howl, slightly less during the cry and much lower during the mew. The cricothyroid intensity was highest during the mew, but much less during the cry and the howl. The cricothyroid frequency spread was the highest between 50 and 300 Hz during the mew, howl, and cry. The genioglossus muscle also had a flat contour for peak frequency versus intensity spread between the three vocals, its intensity was highest for the howl, followed by the cry and the mew. Its frequency spread was between 175 and 300 Hz. Figure 15 illustrates the power density spectrum and frequency shifts of the laryngeal, oral and respiratory muscles for the low (mew) and high intensity (howl) vocalizations. Table 2 provides comparison between the mew, howl and the cry illustrating neither the thyroarytenoid nor the diaphragm activity were significantly different between the howl and the cry. However, the cricothyroid showed clear difference in its activation patterns between the three vocalizations.
      FIGURE 13
      FIGURE 13Activation cascades of the laryngeal, respiratory, and oral muscles indexed to point of vocalization onset.
      FIGURE 14
      FIGURE 14Intensity versus frequency of activation of laryngeal, respiratory, and oral muscles for the mew (M), howl (H), and cry (C). Sharp gradient exhibited by the thyroarytenoid muscle for the mew, cry, and howl can be seen as against normal curves for other muscles.
      FIGURE 15
      FIGURE 15Power density spectrum of laryngeal respiratory and oral muscles for the mew and the howl. A comparison was made between the mew (black) and the howl (red). The cry followed much the same patterns as the howl.

       Peak Intensity

      Figure 16 illustrates the average peak contractile intensity of the laryngeal, tongue, and respiratory muscles during the mew and howl. The abdominal muscles showed the largest intensity of activation for any type of vocalization, with the highest for the external abdominal oblique muscle. In the laryngeal cluster, the thyroarytenoid had a higher peak intensity compared to that of the cricothyroid for the howl and the cry.
      FIGURE 16
      FIGURE 16Histogram illustrating the average peak contractile intensity of the mew, howl, and cry for the laryngeal, respiratory, and tongue muscles.

       Turns and Amplitude Analysis

      The motor unit potentials of cricothyroid and thyroarytenoid muscles were studied during the mew, howl, and cry. The howl was designated as high intensity, the cry as medium, and the mew as low intensity, based on their decibel values. For both muscles, the turns varied between 120 and 550 turns/s while the amplitude varied between 100 and 500 μV for the mew, howl, and cry (Table 3). The cricothyroid muscle exhibited higher value for amplitude for the mew than the thyroarytenoid.
      TABLE 3Turns and Amplitude Values of the Cricothyroid (CT) and Thyroarytenoid (TA) Muscles for the Mew, Howl, and Cry
      HowlCryMew
      Amplitude (µV)Turns per secondAmplitude (µV)Turns per secondAmplitude (µV)Turns per second
      CT380±70420±60350±60310±70260±40180±50
      TA370±40430±50320±40330±50220±50160±40

       Respiratory Frequency, Timing, and Effort Modulation

      During mew episodes the respiratory frequency decreased from 40 ± 4 breaths/min to 32 ± 5 breaths/min (n = 10, P < 0.05). The Ti showed a reduction from 0.60 ± 0.05s to 0.40 ± 0.01s (P < 0.05) while the Te was prolonged, from 0.9 ± 0.05s to an average of 1.5 ± 0.50s (P < 0.05). During the howl, the respiratory frequency decreased from 40 ± 3 breaths/min to 26 ± 5 breaths/min (n = 10, P < 0.05). The Ti showed a reduction from 0.60 ± 0.05s to 0.50 ± 0.01s (P < 0.05) while the Te was prolonged, from 0.9 ± 0.05s to an average of 2.5 ± 0.50s (P < 0.05). The cry had the strongest decrease in breathing function, the respiratory frequency from 40 ± 3 breaths/min to 11 ± 3 breaths/min (n = 10, P < 0.05). The Ti showed a reduction from 0.60 ± 0.05s to 0.50 ± 0.01s (P < 0.05) while the Te was prolonged from 0.9 ± 0.05s to an average of 4.5 ± 0.50s (P < 0.05). The diaphragm EMG amplitude increased up to four times that of eupnea prior to the vocalization episode signifying increases to inspiratory effort. Figure 17 illustrates the percentage change in individual respiratory parameters during the mew, howl, and cry.
      FIGURE 17
      FIGURE 17Net changes to respiratory frequency (RF), inspiratory (Ti), and expiratory (Te) durations for the mew, cry, and howl.

       Cardiovascular Changes During Vocalization

      During vocalization episodes the blood pressure increased from 100 ± 5 mmHg to 120 ± 10 mm HG (n = 10, P < 0.05) representing a 25% increase (Figure 18). The heart rate increased from 199 ± 5 beats/min to 220 ± 5 beats/min (n = 10, P < 0.05) representing a 10% increase. Blood pressure and heart rate returned to eupneic values immediately following the end of vocalization.
      FIGURE 18
      FIGURE 18Blood pressure responses during a vocalization episode.

      DISCUSSION

      The major finding of this study is that positive and negative vocal expressions possess different laryngeal, oral, and respiratory motor patterning as hypothesized. The results show that there exists a close relationship between the laryngeal, respiratory, tongue and jaw muscle activations and the generation of voice. In the following sections, these findings are outlined and discussed how they are related to vocalization mechanisms.

       Description of the Vocalization Types That Can Be Generated in the PAG

      Zhang et al
      • Zhang SP
      • Davis PJ
      • Bandler R
      Brain stem integration of vocalization: role of midbrain periaqueductal gray.
      reported two types of vocalization that can be generated by stimulating in the PAG. They classified them as “voiced” (mew, howl, and growl) and “unvoiced” (hiss) based on the loudness of the sound. The loudness of the sound does not actually describe a PAG-induced mew or howl, but depends on the volume of the microinjection and amplification/sampling during data acquisition. In order to connect the sounds generated in the PAG to specific emotional expressions, their activation patterns are required. In this study, the frequency, decibel levels, and durations of the mew, howl, and cry were calculated. These parameters match the normal mewing, crying, and howling behavior of cats.
      • John ER
      • Chesler P
      • Bartlett F
      • Victor I
      Observation learning in cats.
      • Bradshaw JWS
      The Behavior of the Domestic Cat.
      • Landsberg G
      Feline behavior and welfare.
      • Bradshaw J
      • Cameron-Beaumont C
      The signaling repertoire of the domestic cat and its undomesticated relatives.
      • Beaver BV
      Feline Behavior: A Guide for Veterinarians.
      • Yeon SC
      • Kim YK
      • Park SJ
      Differences between vocalization evoked by social stimuli in feral cats and house cats.

       Topographical Segregation of Vocalization Areas Within the PAG

      The surgical method of exposing the PAG provided visual control of the PAG while implanting the micropipette as compared to stereotaxic approach. The advantage of the excitatory amino acid stimulation technique is that only neuronal cell bodies and the dendritic processes are stimulated and not any fibers of the passage.
      • Fries W
      • Zieglgansberger W
      A method to discriminate axonal from cell body activity and to analyze ‘silent’ cells.
      ,
      • Goodchild AK
      • Dampney RAL
      • Bandler R
      A method for evoking physiological responses by stimulation of cell bodies, but not axons of passage, within localized regions of the central nervous system.
      The vocalization response usually began within 10–15 seconds after the DLH-injection and was strictly limited to specific parts of the PAG and adjacent tegmentum. Mews were evoked from only the lateral column of the PAG, howls from only the ventrolateral column. Cries were evoked from two areas, the rostral and the caudal-ventrolateral PAG. This shows that circuits producing positive and negative emotional vocalizations are separated within the PAG. Hissing, the nonvocal sound,
      • Zhang SP
      • Davis PJ
      • Bandler R
      Brain stem integration of vocalization: role of midbrain periaqueductal gray.
      ,
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      was never been generated as a specific train of vocalization, but only together with the howl.

       Emotional Encoding of the Mew, Cry, and Howl

      Mewing is seen in acts of playfulness, expression of joy and affection.
      • John ER
      • Chesler P
      • Bartlett F
      • Victor I
      Observation learning in cats.
      ,
      • Landsberg G
      Feline behavior and welfare.
      In monkeys, chucking and cackling calls are described as expressions of joy, laughter, and playfulness. They also have been elicited in the lateral PAG.
      • Jürgens U
      The neural control of vocalization in mammals: a review.
      In humans, activation of the PAG is seen in a wide range of positive emotional states such as unconditional love and pleasure.
      • Linnman C
      • Moulton EA
      • Barmettler G
      • Becerra L
      • Borsook D
      Neuroimaging of the periaqueductal gray: state of the field.
      The lateral PAG maintains reciprocal connections with the lateral part of the pontine parabrachial nucleus
      • Krout KE
      • Jansen AS
      • Loewy AD
      Periaqueductal gray matter projection to the parabrachial nucleus in rat.
      • Bernard JF
      • Dallel R
      • Raboisson P
      • Villanueva L
      • Le Bars D
      Organization of the efferent projections from the spinal cervical enlargement to the parabrachial area and periaqueductal gray: a PHA-L study in the rat.
      • Wiberg M
      Reciprocal connections between the periaqueductal gray matter and other somatosensory regions of the cat midbrain: a possible mechanism of pain inhibition.
      • Holstege G
      Descending motor pathways and the spinal motor system: limbic and non-limbic components.
      • Holstege G
      Descending pathways from the periaqueductal gray and adjacent areas.
      • Carrive P
      The periaqueductal gray and defensive behavior: functional representation and neuronal organization.
      ; which is involved in a large number of homeostatic functions. Mewing is also an emotional expression of food seeking behavior.
      • John ER
      • Chesler P
      • Bartlett F
      • Victor I
      Observation learning in cats.
      • Bradshaw JWS
      The Behavior of the Domestic Cat.
      • Landsberg G
      Feline behavior and welfare.
      • Bradshaw J
      • Cameron-Beaumont C
      The signaling repertoire of the domestic cat and its undomesticated relatives.
      • Beaver BV
      Feline Behavior: A Guide for Veterinarians.
      • Yeon SC
      • Kim YK
      • Park SJ
      Differences between vocalization evoked by social stimuli in feral cats and house cats.
      Thus, the present finding that mewing is encoded in the lateral PAG fits the concept that this part of the PAG is specifically involved in the motor expression of positive emotions.
      On the other hand, howling is a strong feline vocal expression for communication of fear, danger, anxiety, and/or medical/emotional disturbance.
      • John ER
      • Chesler P
      • Bartlett F
      • Victor I
      Observation learning in cats.
      ,
      • Landsberg G
      Feline behavior and welfare.
      This was evoked only from the ventrolateral PAG previously shown to be involved in mediation of fear and anxiety.
      • Bandler R
      • Carrive P
      Integrated defence reaction elicited by excitatory amino acid injection in the midbrain periaqueductal gray region of the unrestrained cat.
      ,
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      ,
      • Subramanian HH
      • Holstege G
      The midbrain periaqueductal gray changes the eupneic respiratory rhythm into a breathing pattern necessary for survival of the individual and of the species.
      ,
      • Beauchaine TP
      Respiratory sinus arrhythmia: a transdiagnostic biomarker of emotion dysregulation and psychopathology.
      Respiratory sinus arrhythmia, a motor signature of strong fear or anxiety reaction,
      • Krukoff TL
      • Harris KH
      • Jhamandas JH
      Efferent projections from the parabrachial nucleus demonstrated with the anterograde tracer Phaseolus vulgaris leucoagglutinin.
      was observed during howling. In monkeys, the “trill” has been suggested to express anger, fear, and/or anxiety
      • Jürgens U
      The neural control of vocalization in mammals: a review.
      and may be comparable with the howl in the cat. Although precise topography of the trill in the monkey PAG is not known, neuronal firing during the trill was recorded in the ventrolateral PAG.
      • Düsterhöft F
      • Häusler U
      • Jürgens U
      Neuronal activity in the periaqueductal gray and bordering structures during vocal communication in the squirrel monkey.
      The present data showing that howling is encoded in the ventrolateral PAG fits the concept that this part of the PAG is involved in the motor expression of negative emotions.
      Generation of crying in the dorsal and the caudal-ventrolateral PAG has not been shown before. The dorsal and ventrolateral PAG maintains dense connections with the analgesic system in the brainstem,
      • Holstege G
      Descending pathways from the periaqueductal gray and adjacent areas.
      ,
      • Sukhotinsky I
      • Reiner K
      • Govrin-Lippmann R
      • Belenky M
      • Lu J
      • Hopkins DA
      • Saper CB
      • Devor M
      Projections from the mesopontine tegmental anesthesia area to regions involved in pain modulation.
      • Behbehani MM
      Functional characteristics of the midbrain periaqueductal gray.
      • Li YQ
      • Shinonaga Y
      • Takada M
      • Mizuno N
      Demonstration of axon terminals of projection fibers from the periaqueductal gray onto neurons in the nucleus raphe magnus which send their axons to the trigeminal sensory nuclei.
      • Holstege G
      Direct and indirect pathways to lamina I in the medulla oblongata and spinal cord of the cat.
      • Cowie RJ
      • Holstege G
      Dorsal mesencephalic projections to pons, medulla, and spinal cord in the cat: limbic and non-limbic components.
      • Wiberg M
      • Westman J
      • Blomqvist A
      Somatosensory projection to the mesencephalon: an anatomical study in the monkey.
      • Hayashi H
      • Sumino R
      • Sessle BJ
      Functional organization of trigeminal subnucleus interpolaris: nociceptive and innocuous afferent inputs, projections to thalamus, cerebellum, and spinal cord, and descending modulation from periaqueductal gray.
      • Mantyh PW
      The ascending input to the midbrain periaqueductal gray of the primate.
      while the caudal-ventrolateral PAG is involved in fear processing.
      • Darwin CR
      The expression of the emotions in man and animals.
      ,
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      ,
      • Subramanian HH
      • Holstege G
      The midbrain periaqueductal gray changes the eupneic respiratory rhythm into a breathing pattern necessary for survival of the individual and of the species.
      ,
      • Beauchaine TP
      Respiratory sinus arrhythmia: a transdiagnostic biomarker of emotion dysregulation and psychopathology.
      The finding that the cries were generated from this region fits the concept that this PAG region is involved in the expression of pain and/or fear or signaling emotional problems.
      In the present study, growling was never evoked. Growling reported by Bandler and Carrive
      • Bandler R
      • Carrive P
      Integrated defence reaction elicited by excitatory amino acid injection in the midbrain periaqueductal gray region of the unrestrained cat.
      was based on stimulation in the PAG in chronically implanted freely moving animals. In these animals growl was evoked as a component of rage reaction. In the present study, a rage or aggressive reaction from the PAG in precollicular decerebrate preparations was never observed.
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      ,
      • Subramanian HH
      • Holstege G
      The midbrain periaqueductal gray changes the eupneic respiratory rhythm into a breathing pattern necessary for survival of the individual and of the species.
      Perhaps an intact brain is required for expression of the growl. Although growling could well be encoded in the PAG, its elicitation could be a higher brain function.

       The Need for Quantitative Laryngeal Electromyography

      Validation of quantitative EMGs provides standardization for understanding laryngeal motor control.
      • Statham MM
      • Rosen CA
      • Nandedkar SD
      • Munin MC
      Quantitative laryngeal electromyography: turns and amplitude analysis.
      ,
      • Heman-Ackah YD
      • Mandel S
      • Manon-Espaillat R
      • Abaza MM
      • Sataloff RT
      Laryngeal electromyography.
      ,
      • Blitzer A
      • Crumley RI
      • Dailey SH
      Recommendations of the neurolaryngology study group on laryngeal electromyography.
      Four different types of quantitative analysis of laryngeal, respiratory, jaw, and tongue muscles were executed during the mew, howl, and cry. They were activation constants, intensity versus frequency of activation, power density spectrum and analysis of turns, and amplitude, providing precise activity patterns of each muscle.

       The Role of Cricothyroid, Thyroarytenoid, and Posterior Cricoarytenoid in Vocal Expression

      In humans, the evaluation of vocal-fold tension is primarily via qualitative laryngoscopy and stroboscopy.
      • Heman-Ackah YD
      • Mandel S
      • Manon-Espaillat R
      • Abaza MM
      • Sataloff RT
      Laryngeal electromyography.
      ,
      • Wu AP
      • Sulica L
      Diagnosis of vocal fold paresis: current opinion and practice.
      It is known that in humans, the thyroarytenoid muscle adducts the vocal folds to create vibration, the PCA muscle abducts the vocal folds for opening the glottal space while the cricothyroid muscle is considered to be the laryngeal tensor. Nevertheless, conclusive evidence is still lacking on which of these three muscles is the primary determinant and which is the modulator of the vocal fold tension.
      • Heman-Ackah YD
      • Mandel S
      • Manon-Espaillat R
      • Abaza MM
      • Sataloff RT
      Laryngeal electromyography.
      ,
      • Wu AP
      • Sulica L
      Diagnosis of vocal fold paresis: current opinion and practice.
      ,
      • Titze IR
      • Jiang J
      • Drucker DG
      Preliminaries to the body-cover theory of pitch control.
      The present EMG results indicate that the cricothyroid muscle is the primary determinant of the gross vocal fold tension, because it was always time-locked with the on- and offset of vocalization and did not exhibit any frequency shift between the mew, cry and howl. The cricothyroid also exhibited the highest values for turns and amplitude for all the three vocals. The cricothyroid, therefore, generates the vocal fold tension.
      The thyroarytenoid plays a modulatory role, because it showed a normal gradient for all three vocals on intensity versus peak frequency of activation, and did not exhibit any specific synchrony with other laryngeal or respiratory muscles during vocal production. It exhibited pre- and postvocalization activities, of which the shapes, intensities, and extents were different during mew, howl, and cry. During the cry the activation pattern of the thyroarytenoid was very different to its activation pattern during mew or howl, because it exhibited a decrementing discharge pattern. Its activity started even before the actual crying. The turns and amplitude values for the thyroarytenoid were also lower than those of the cricothyroid for all three vocals. These findings indicate that the thyroarytenoid plays a modulatory role in vocal fold tension, leading to differences in pitch and loudness of the sound. The major finding is that the increased muscular force required for generating the howl and cry (as compared to the mew) is brought about by an increase in firing frequency of the motor units of the thyroarytenoid. This is similar to what has been theorized in humans
      • Titze IR
      • Jiang J
      • Drucker DG
      Preliminaries to the body-cover theory of pitch control.
      and observed during tasks that involved different intensity of phonation.
      • Lindestad PA
      • Fritzell B
      • Persson A
      Quantitative analysis of laryngeal EMG in normal subjects.
      The PCA was inhibited during vocal expression. Its preinspiratory activation suggests that it plays a role in glottal control rather than producing sound. Investigation in dogs and humans suggests that it functions as the sole abductor of the glottis, not only during phonation but also during coughing
      • Chhetri DK1
      • Neubauer J
      • Sofer E
      Posterior cricoarytenoid muscle dynamics in canines and humans.
      and swallowing.
      • Ota R
      • Takakusaki K
      • Katada A
      • Harada H
      • Nonaka S
      • Harabuchi Y
      Contribution of the lateral lemniscus to the control of swallowing in decerebrate cats.

       Synchrony Between the Three Motor Systems During Vocal Expression

      Another important finding is that the laryngeal system interacts with the respiratory system during vocal production. Similar interaction can be seen in other systems such as swallowing and coughing.
      • Bolser DC
      • Gestreau C
      • Morris KF
      • Davenport PW
      • Pitts TE
      Central neural circuits for coordination of swallowing, breathing, and coughing: predictions from computational modeling and simulation.
      The PAG also produces a range of respiratory behaviors independent of vocal behavior,
      • Subramanian HH
      • Balnave RJ
      • Holstege G
      The midbrain periaqueductal gray control of respiration.
      ,
      • Subramanian HH
      • Holstege G
      The midbrain periaqueductal gray changes the eupneic respiratory rhythm into a breathing pattern necessary for survival of the individual and of the species.
      ,
      • Subramanian HH
      • Holstege G
      Midbrain and medullary control of postinspiratory activity of the crural and costal diaphragm in vivo.
      including modulation of key respiratory neurons that are involved in rhythm generation.
      • Subramanian HH
      • Holstege G
      Stimulation of the midbrain periaqueductal gray modulates preinspiratory neurons in the ventrolateral medulla in the rat in vivo.
      ,
      • Subramanian HH
      Descending control of the respiratory neuronal network by the midbrain periaqueductal grey in the rat in vivo.
      However, during emotional vocalization in mammals, the vocal system drives the respiratory system, which was also shown in songbirds.
      • Schmidt MF
      • Wild JM
      The respiratory-vocal system of songbidrds: anatomy, physiology and neural control.
      Irrespective of the type of vocalization, the internal and external abdominal expiratory muscles were the first to be activated during the vocal expiratory phase, followed by the thyroarytenoid and genioglossus and finally the cricothyroid, which was time-locked with the vocalization onset. The pressure, volume and velocity of airflow required for vocal production seemed to be exquisitely controlled by the synchrony between the thyroarytenoid, genioglossus, and the abdominal muscles. This was also evident from the tracheal pressure generated for all three vocalizations. The diaphragm showed a range of inspiratory efforts, the lowest for the mew and the highest for the howl. Interestingly, activation of the diaphragm was also found during long vocal outflow such as the cry. A high frequency shift of the diaphragm and the internal intercostal muscles were seen during the howl, which might signify fatigue.
      • Janssens L
      • Brumagne S
      • McConnell AK
      • Raymaekers J
      • Goossens N
      • Gayan-Ramirez G
      • Hermans G
      • Troosters T
      The assessment of inspiratory muscle fatigue in healthy individuals: a systematic review.
      • Road JD
      • Jiang TX
      Determinants of diaphragmatic injury.
      • Cairns AM
      • Road JD
      High-frequency oscillation and centroid frequency of diaphragm EMG during inspiratory loading.
      • Chen R
      • Collins SJ
      • Remtulla H
      • Parkes A
      • Bolton CF
      Needle EMG of the human diaphragm: power spectral analysis in normal subjects.
      This was also found during the inspiratory muscle fatigue noticed in humans during classical singing, involving high pitch and sustained vocal outflow.
      • Pettersen V
      Muscular patterns and activation levels of auxiliary breathing muscles and thorax movement in classical singing.
      In cat fatigue was seen in only the expiratory intercostal but not in the abdominal muscles, while abdominal expiratory fatigue was found during classical singing in humans.
      • Pettersen V
      Muscular patterns and activation levels of auxiliary breathing muscles and thorax movement in classical singing.
      Both the genioglossus and digastric muscles were synchronized with the thyroarytenoid. Quantitative analysis revealed that the active tension component of the thyroarytenoid was higher than that of the genioglossus and digastric muscles during howling and crying. Apparently, the thyroarytenoid muscle functions as a fine tensor of the vocal fold as in humans during vocalizations of longer duration and higher pitch.
      • Johns MM
      • Urbanchek M
      • Chepeha DB
      • Kuzon Jr., WM
      • Hogikyan ND
      Length-tension relationship of the feline thyroarytenoid muscle.

       Animal Model Control Data for Diagnosis of Speech Disorders

      According to some reports
      • Nash EA
      • Ludlow CL
      Laryngeal muscle activity during speech breaks in adductor spasmodic dysphonia.
      ,
      • Cyrus CB
      • Bielamowicz S
      • Evans FJ
      • Ludlow CL
      Adductor muscle activity abnormalities in abductor spasmodic dysphonia.
      only the thyroarytenoid muscle is affected in adductor spasmodic dysphonia, while other studies
      • Broniatowski M
      • Grundfest-Broniatowski S
      • Hahn EC
      • Hadley AJ
      • Tyler DJ
      • Tucker HM
      Selective intraoperative stimulation of the human larynx.
      ,
      • Watson BC
      • Schaefer SD
      • Freeman FJ
      • Dembowski J
      • Kondraske G
      • Roark R
      Laryngeal electromyographic activity in adductor and abductor spasmodic dysphonia.
      convey that both the thyroarytenoid and PCA are affected. In Parkinson's disease with common speech motor deficits, a variety of abnormal laryngeal muscle activation patterns are seen.
      • Tawadros PB
      • Cordato D
      • Cathers I
      • Burne JA
      An electromyographic study of parkinsonian swallowing and its response to levodopa.
      However, there exists no general consensus regarding the laryngeal muscle abnormalities integral to speech disorders.
      • Blitzer A
      • Crumley RI
      • Dailey SH
      Recommendations of the neurolaryngology study group on laryngeal electromyography.
      Perhaps the present data concerning the various muscle activation patterns and their quantification might serve as a valuable model for a diagnostic assessment of the laryngeal and upper airway disorders in humans.

      CONCLUSIONS

      The present results show that positive and negative vocal expressions are generated by different neural circuits within the PAG. These circuits enable specific activation patterns of laryngeal, respiratory, tongue, and jaw muscles for the mew, howl, hiss, and cry vocal expressions. The data also show that stimulation of premotor interneurons in the NRA projecting to the laryngeal and respiratory muscle motoneurons never produce similar emotional vocalizations, only various guttural sounds, which fits the concept that the PAG uses the NRA as a tool for producing vocalizations.

      Acknowledgment

      The advise on spectral, statistical and quantitative EMG and vocalization data analysis by Dr Roger Adams, Dr. Andrew Bradley, and Dr. Chin Moi Chow are gratefully acknowledged.

      Conflicts of interest

      The authors declare no conflict of interests.

      Author contributions

      The animal experiments were performed in accordance with appropriate ethics approval at The University of Sydney between 1997 and 2000. HHS, RJB and GH conceived and designed the project. HHS performed the experiments. GH and RJB supervised the experiments. HHS, RJB and GH analyzed primary data. HHS and GH made figure illustrations. RJB checked and verified figure illustrations. GH designed and curated the final data and figure illustrations representation. HHS and GH interpreted data and wrote the paper.

      Appendix. SUPPLEMENTARY DATA

      REFERENCES

        • John ER
        • Chesler P
        • Bartlett F
        • Victor I
        Observation learning in cats.
        Science. 1968; 159: 1489-1492
        • Bradshaw JWS
        The Behavior of the Domestic Cat.
        Redwood Press, Bristol1992
        • Landsberg G
        Feline behavior and welfare.
        J Am Vet Med Assoc. 1996; 208: 502-505
        • Bradshaw J
        • Cameron-Beaumont C
        The signaling repertoire of the domestic cat and its undomesticated relatives.
        in: Turner DC Bateson PPG The Domestic Cat: The Biology of Its Behavior. Cambridge University Press, Cambridge2000: 71-72
        • Beaver BV
        Feline Behavior: A Guide for Veterinarians.
        Elsevier Science, St. Louis2003
        • Yeon SC
        • Kim YK
        • Park SJ
        Differences between vocalization evoked by social stimuli in feral cats and house cats.
        Behav Process. 2011; 87: 183-189
        • Nicastro N
        • Owren MJ
        Classification of domestic cats (felis catus) vocalizations by naïve and experienced human listeners.
        J Comp Psychol. 2003; 117: 44-52
        • Darwin CR
        The expression of the emotions in man and animals.
        John Murray. 1st edition, London1872
        • Bandler R
        • Carrive P
        Integrated defence reaction elicited by excitatory amino acid injection in the midbrain periaqueductal gray region of the unrestrained cat.
        Brain Res. 1988; 439: 95-106
        • Carrive P
        • Dampney RA
        • Bandler R
        Somatic and autonomic integration in the midbrain of the unanesthetized decerebrate cat: a distinctive pattern evoked by excitation of neurones in the subtentorial portion of the midbrain periaqueductal grey.
        Brain Res. 1989; 483: 251-258
        • Zhang SP
        • Davis PJ
        • Bandler R
        Brain stem integration of vocalization: role of midbrain periaqueductal gray.
        J Neurophysiol. 1994; 72: 1-19
        • Subramanian HH
        • Balnave RJ
        • Holstege G
        The midbrain periaqueductal gray control of respiration.
        J Neurosci. 2008; 28: 12274-12283
        • Subramanian HH
        • Holstege G
        The midbrain periaqueductal gray changes the eupneic respiratory rhythm into a breathing pattern necessary for survival of the individual and of the species.
        Prog Brain Res. 2014; 212: 352-384
        • Skultety FM
        The behavioral effects of destructive lesions of the periaqueductal gray matter in adult cats.
        J Comp Neurol. 1958; 110: 337-365
        • Adametz J
        • O'Leary JL
        Experimental mutism resulting from periaqueductal lesions in cats.
        Neurology. 1959; 9: 636-642
        • Holstege G
        Anatomical study of final common pathway for vocalization in the cat.
        J Comp Neurol. 1989; 284: 242-252
        • Alexander MP
        Chronic akinetic mutism after mesencephalic-diencephalic infarction: remediated with dopaminergic medications.
        Neurorehabil Neural Repair. 2001; 15: 151-156
        • Holstege G
        • Subramanian HH
        Two different motor systems generate human speech.
        J Comp Neurol. 2016; 524: 1558-1577
        • Zhang SP
        • Davis PJ
        • Bandler R
        Brain stem integration of vocalization: role of nucleus retroambigualis.
        J Neurophysiol. 1996; 74: 1-13
        • Subramanian HH
        • Holstege G
        The nucleus retroambiguus control of respiration.
        J Neurosci. 2009; 29: 3824-3832
        • Subramanian HH
        • Holstege G
        Midbrain and medullary control of postinspiratory activity of the crural and costal diaphragm in vivo.
        J Neurophysiol. 2011; 105: 2852-2862
        • Fries W
        • Zieglgansberger W
        A method to discriminate axonal from cell body activity and to analyze ‘silent’ cells.
        Exp Brain Res. 1974; 21: 441-445
        • Goodchild AK
        • Dampney RAL
        • Bandler R
        A method for evoking physiological responses by stimulation of cell bodies, but not axons of passage, within localized regions of the central nervous system.
        J Neurosci Methods. 1982; 6: 351-363
        • Berman AL
        The Brainstem of the Cat. A Cytoarchitectonic Atlas with Stereotaxic Coordinates.
        University of Wisconsin Press, Madison, WI1968
        • Statham MM
        • Rosen CA
        • Nandedkar SD
        • Munin MC
        Quantitative laryngeal electromyography: turns and amplitude analysis.
        Laryngoscope. 2010; 120: 2036-2041
        • Jürgens U
        The neural control of vocalization in mammals: a review.
        J Voice. 2009; 23: 1-10
        • Linnman C
        • Moulton EA
        • Barmettler G
        • Becerra L
        • Borsook D
        Neuroimaging of the periaqueductal gray: state of the field.
        Neuroimage. 2012; 60: 505-522
        • Krout KE
        • Jansen AS
        • Loewy AD
        Periaqueductal gray matter projection to the parabrachial nucleus in rat.
        J Comp Neurol. 1998; 401: 437-454
        • Bernard JF
        • Dallel R
        • Raboisson P
        • Villanueva L
        • Le Bars D
        Organization of the efferent projections from the spinal cervical enlargement to the parabrachial area and periaqueductal gray: a PHA-L study in the rat.
        J Comp Neurol. 1995; 353: 480-505
        • Wiberg M
        Reciprocal connections between the periaqueductal gray matter and other somatosensory regions of the cat midbrain: a possible mechanism of pain inhibition.
        Ups J Med Sci. 1992; 97: 37-47
        • Holstege G
        Descending motor pathways and the spinal motor system: limbic and non-limbic components.
        Prog Brain Res. 1991; 87: 307-421
        • Holstege G
        Descending pathways from the periaqueductal gray and adjacent areas.
        The Midbrain Periaqueductal Gray Matter, Functional, Anatomical and Neurochemical Organization. Plenum, New York, NY1991: 239-265
        • Carrive P
        The periaqueductal gray and defensive behavior: functional representation and neuronal organization.
        Behav Brain Res. 1993; 58: 27-47
        • Beauchaine TP
        Respiratory sinus arrhythmia: a transdiagnostic biomarker of emotion dysregulation and psychopathology.
        Curr Opin Psychol. 2015; 23: 43-47
        • Krukoff TL
        • Harris KH
        • Jhamandas JH
        Efferent projections from the parabrachial nucleus demonstrated with the anterograde tracer Phaseolus vulgaris leucoagglutinin.
        Brain Res Bull. 1993; 30: 163-172
        • Düsterhöft F
        • Häusler U
        • Jürgens U
        Neuronal activity in the periaqueductal gray and bordering structures during vocal communication in the squirrel monkey.
        Neuroscience. 2004; 123: 53-60
        • Sukhotinsky I
        • Reiner K
        • Govrin-Lippmann R
        • Belenky M
        • Lu J
        • Hopkins DA
        • Saper CB
        • Devor M
        Projections from the mesopontine tegmental anesthesia area to regions involved in pain modulation.
        J Chem Neuroanat. 2006; 32: 159-178
        • Behbehani MM
        Functional characteristics of the midbrain periaqueductal gray.
        Prog Neurobiol. 1995; 46: 575-605
        • Li YQ
        • Shinonaga Y
        • Takada M
        • Mizuno N
        Demonstration of axon terminals of projection fibers from the periaqueductal gray onto neurons in the nucleus raphe magnus which send their axons to the trigeminal sensory nuclei.
        Brain Res. 1993; 608: 138-140
        • Holstege G
        Direct and indirect pathways to lamina I in the medulla oblongata and spinal cord of the cat.
        Progr Brain Res. 1988; 77: 47-94
        • Cowie RJ
        • Holstege G
        Dorsal mesencephalic projections to pons, medulla, and spinal cord in the cat: limbic and non-limbic components.
        J Comp Neurol. 1992; 319: 536-559
        • Wiberg M
        • Westman J
        • Blomqvist A
        Somatosensory projection to the mesencephalon: an anatomical study in the monkey.
        J Comp Neurol. 1987; 264: 92-117
        • Hayashi H
        • Sumino R
        • Sessle BJ
        Functional organization of trigeminal subnucleus interpolaris: nociceptive and innocuous afferent inputs, projections to thalamus, cerebellum, and spinal cord, and descending modulation from periaqueductal gray.
        J Neurophysiol. 1984; 51: 890-905
        • Mantyh PW
        The ascending input to the midbrain periaqueductal gray of the primate.
        J Comp Neurol. 1982; 211: 50-64
        • Heman-Ackah YD
        • Mandel S
        • Manon-Espaillat R
        • Abaza MM
        • Sataloff RT
        Laryngeal electromyography.
        Otolaryngol Clin North Am. 2007; 40: 1003-1023
        • Blitzer A
        • Crumley RI
        • Dailey SH
        Recommendations of the neurolaryngology study group on laryngeal electromyography.
        Otolaryngol Head Neck Surg. 2009; 140: 782-793
        • Wu AP
        • Sulica L
        Diagnosis of vocal fold paresis: current opinion and practice.
        Laryngoscope. 2015; 125: 904-908
        • Titze IR
        • Jiang J
        • Drucker DG
        Preliminaries to the body-cover theory of pitch control.
        J Voice. 1988; 1: 314-319
        • Lindestad PA
        • Fritzell B
        • Persson A
        Quantitative analysis of laryngeal EMG in normal subjects.
        Acta Otolaryngol. 1991; 111: 1146-1152
        • Chhetri DK1
        • Neubauer J
        • Sofer E
        Posterior cricoarytenoid muscle dynamics in canines and humans.
        Laryngoscope. 2014; 124: 2363-2367
        • Ota R
        • Takakusaki K
        • Katada A
        • Harada H
        • Nonaka S
        • Harabuchi Y
        Contribution of the lateral lemniscus to the control of swallowing in decerebrate cats.
        Neuroscience. 2013; 254: 260-274
        • Bolser DC
        • Gestreau C
        • Morris KF
        • Davenport PW
        • Pitts TE
        Central neural circuits for coordination of swallowing, breathing, and coughing: predictions from computational modeling and simulation.
        Otolaryngol Clin North Am. 2013; 46: 957-964
        • Subramanian HH
        • Holstege G
        Stimulation of the midbrain periaqueductal gray modulates preinspiratory neurons in the ventrolateral medulla in the rat in vivo.
        J Comp Neurol. 2013; 521: 3083-3093
        • Subramanian HH
        Descending control of the respiratory neuronal network by the midbrain periaqueductal grey in the rat in vivo.
        J Physiol. 2013; 591: 109-122
        • Schmidt MF
        • Wild JM
        The respiratory-vocal system of songbidrds: anatomy, physiology and neural control.
        Progress Brain Res. 2014; 212: 297-335
        • Janssens L
        • Brumagne S
        • McConnell AK
        • Raymaekers J
        • Goossens N
        • Gayan-Ramirez G
        • Hermans G
        • Troosters T
        The assessment of inspiratory muscle fatigue in healthy individuals: a systematic review.
        Respir Med. 2013; 107: 331-346
        • Road JD
        • Jiang TX
        Determinants of diaphragmatic injury.
        Mol Cell Biochem. 1998; 179: 81-86
        • Cairns AM
        • Road JD
        High-frequency oscillation and centroid frequency of diaphragm EMG during inspiratory loading.
        Respir Physiol. 1998; 112: 305-313
        • Chen R
        • Collins SJ
        • Remtulla H
        • Parkes A
        • Bolton CF
        Needle EMG of the human diaphragm: power spectral analysis in normal subjects.
        Muscle Nerve. 1996; 19: 324-330
        • Pettersen V
        Muscular patterns and activation levels of auxiliary breathing muscles and thorax movement in classical singing.
        Folia Phoniatr Logop. 2005; 57: 255-277
        • Johns MM
        • Urbanchek M
        • Chepeha DB
        • Kuzon Jr., WM
        • Hogikyan ND
        Length-tension relationship of the feline thyroarytenoid muscle.
        J Voice. 2004; 18: 285-291
        • Nash EA
        • Ludlow CL
        Laryngeal muscle activity during speech breaks in adductor spasmodic dysphonia.
        Laryngoscope. 1996; 106: 484-489
        • Cyrus CB
        • Bielamowicz S
        • Evans FJ
        • Ludlow CL
        Adductor muscle activity abnormalities in abductor spasmodic dysphonia.
        Otolaryngol Head Neck Surg. 2001; 124: 23-30
        • Broniatowski M
        • Grundfest-Broniatowski S
        • Hahn EC
        • Hadley AJ
        • Tyler DJ
        • Tucker HM
        Selective intraoperative stimulation of the human larynx.
        Laryngoscope. 2012; 122: 2015-2022
        • Watson BC
        • Schaefer SD
        • Freeman FJ
        • Dembowski J
        • Kondraske G
        • Roark R
        Laryngeal electromyographic activity in adductor and abductor spasmodic dysphonia.
        J Speech Hear Res. 1991; 34: 473-482
        • Tawadros PB
        • Cordato D
        • Cathers I
        • Burne JA
        An electromyographic study of parkinsonian swallowing and its response to levodopa.
        Mov Disord. 2012; 27: 1811-1815