The Vault

Neural Control of Vocalization & Speech

Pamela J. Davis, Shi Ping Zhang, and Richard Bandler

School of Communication Disorders (P.J.D.) and Department of Anatomy and Histology (S.P.Z., R.B), The University of Sydney, N.S.W., Australia


Although speech is uniquely human, taking half a decade or so to develop, the vocal expression of emotion (e.g. crying, moaning, laughing), occurs in many species and is well developed at, or shortly after, birth. In spite of these developmental differences, both speech and emotional expression are finely controlled by the central nervous system and depend upon the co-ordinated activity of about 80 respiratory, laryngeal (voice box) and oro-facial muscles.

The laryngeal and respiratory muscles and structures of many non-human mammalian species are remarkably similar to that of humans. Thus, whether for speech or vocal expression, and in all mammalian species that vocalize, laryngeal muscles are stretched across the airway and set into vibration (sound production) by airflow resulting from increased expiratory effort in intercostal and abdominal muscles. Further, the timing of the activation and inhibition of different muscle groups is co-ordinated with breathing, and modified to suit the nature of the vocal expression. For example, a single breath may be 10 or even 20 seconds long during speech and song, but is rarely longer than a couple of seconds in quiet breathing. How does the brain produce and control voiced sound?

The Midbrain and the Gray Matter

Research over the past 20 years has shown that a region of the midbrain, the periaqueductal gray matter (PAG), which surrounds the central cerebral aqueduct, is critical for sound production and probably speech, although this region is difficult to study in humans. Lesions of the PAG region have been associated with muteness and the inability to evoke vocalization from other brain regions (for review see Larson, 1988).

We used an animal model of vocalization to determine whether PAG neurons integrate the activity and timing of the respiratory, laryngeal and oral muscles for vocal expression. A model of natural-sounding vocalization was established using excitatory amino acid microinjections (D-L homocysteic acid) in the decerebrate cat, a procedure carried out under deep surgical anaesthesia. This technique specifically excites neuronal cell bodies. Two general types of vocalization, classified as 'voiced' and 'unvoiced', were evoked from neurons located lateral to the cerebral aqueduct in the intermediate part of the PAG. Individual muscles were never activated by neuronal excitation, suggesting that muscle patterns for vocalization are represented in the PAG. Moreover, the timing and muscle activation patterns were strikingly similar to that previously reported for human vowel and voiceless consonant phonation, leading to the conclusion that patterned muscle activity, corresponding to the major categories of voiced and voiceless sound production, are represented in neurons located in the lateral PAG.

Neural pathways from PAG to the respiratory, laryngeal and oro-facial motoneurons in the spinal cord and medulla have yet to be delineated (the motoneurons are the nerve cells to the various muscles). In a second set of experiments, excitation of neurons in the nucleus retroambigualis (NRA) in the medulla (lower part of the brainstem) evoked activity in the respiratory and laryngeal muscles, sometimes with vocalization, but neither functional muscle patterns nor natural sound qualities were produced. Medullary transections which eliminated the NRA also prevented laryngeal motor activity being evoked from the lateral PAG. This indicates that the neural pathway from the PAG to the laryngeal motoneurons includes a path in the caudal medulla, presumably via a synapse in the NRA. The latter is probably important in controlling the laryngeal and respiratory components of sound production, but not the oro-facial modulation of that sound.

A longitudinal section through the mid-brain of the brain. X marks the area of the Periaqueductal Grey Area (PAG) in the brainstem.

PAG, Laryngeal Motor Patterns and Emotional Expression

These and other findings have led to a recent proposal (Zhang et al., 1994, Davis et al., 1995a; 1995b) that neurons in the lateral PAG column could play an important role, not only in the production of emotional vocalization, but also as a generator of specific respiratory and laryngeal motor patterns essential for human speech and song. We suggest that the voluntary motor pathway for the sound production component of speech and song includes higher brain structures which project significantly to the PAG (Jurgens and von Cramon, 1982; Bandler and Shipley, 1994) and excites neuronal networks which coordinate the various muscle patterns for sound production. In contrast to the fine motor control of oral structures, such as the tongue, voluntary control of the larynx and palate is relatively poor when sound is not produced although there can be quite exquisite control of the way the vocal folds are controlled to alter the pitch and quality of a sound. This is consistent with the notion that the voluntary control systems for vocalization are tightly linked with neural circuits such as PAG, which are capable of activating patterned muscle activities.

It is possible that language and speech development is associated with achieving precise voluntary control over the respiratory and laryngeal muscles, and learning to coordinate them with the oro-facial muscles for the production of the great variety of speech sounds. That the PAG region is now considered to be closely involved in the elaboration of emotional experience and expression (Bandler, 1988; Holstege, 1990; Bandler and Shipley, 1994) may explain, in part, the emotional quality of all speech and song.


Supported by grants to: P.D. from the Australian National Health and Medical Research Council; and to R.B. from the Australian National Health and Medical Research Council and the C. and V. Ramaciotti Foundation.


Bandler, R. (1988) "Brain mechanisms of aggression as revealed by electrical and chemical stimulation: Suggestion of a central role for the midbrain periaqueductal grey region". In: Progress in Psychobiology and Physiological Psychology, Vol. 13, edited by A. Epstein and A. Morrison, pp. 67-153, Academic Press.

Bandler, R. and Shipley, M.T. (1994) "Columnar organization in the midbrain periaqueductal gray: Modules for emotional expression? " T.I.N.S. 17: pp.379-389.

Davis, P.J., Winkworth, A., Zhang, S.P and Bandler, R., (1995a) "The neural control of vocalization: Respiratory and emotional influences", J. Voice, in press.

Davis, P.J., Zhang, S.P. and Bandler, R. (1995b) "Midbrain and medullary control of respiration and vocalization, Progress in Brain Research: The Emotional Motor System," G. Holstege, R. Bandler and C. Saper (eds.) (in press)

Holstege, G. (1990) "Subcortical limbic system projections to caudal brainstem and spinal cord." In G. Paxinos, ed., The Human Nervous System. Acad. Press: Tokyo. pp.261-286.

Jurgens, U. and von Cramon, D. (1982) "On the role of the anterior cingulate cortex in phonation: A case report." Brain Lang., 15 pp.234-248.

Larson, C.R. (1988) "Brain mechanisms involved in the control of vocalization." J. Voice, 2: pp.301-311.

Zhang, S.P., Davis, P.J., Bandler, R. and Carrive, P. (1994) "Brain stem integration of vocalization: Role of the midbrain periaqueductal gray." J. Neurophysiol. 72: pp.1337-1356.


Pamela Davis trained as a speech pathologist, graduating in 1964, and has worked mostly with clients with voice disorders, setting up the Voice Clinic at St Vincent's Hospital, Sydney in 1977. In order to understand, more about how the brain produces and controls voice, Pam commenced study of the physiology of the larynx in 1979 in the School of Physiology and Pharmacology at the University of NSW, graduating with a PhD in 1987. In 1986 she was appointed as Senior Lecturer in the School of Communication Disorders at Cumberland College of Health Sciences, now the Faculty of Health Sciences of The University of Sydney. Pam teaches voice science and disorders to undergraduate students of speech pathology, and works with research students interested in voice from various backgrounds— speech pathology, medicine, singing and acting. Pam has published extensively on the physiology of voice and has been an invited or keynote speaker at a number of international conferences over the past 10 years. In 1995 she convened the 9th Vocal Fold Physiology Symposium at The Sydney Opera House for the Voice Foundation (USA).

In 1995 Pam was seconded to a part time position within the Faculty of Health Sciences of The University of Sydney to establish a voice centre. The focus of this unique Australian centre will be its multi disciplinary nature of voice research, care, training and education.


  Bookmark and Share