The different human nonverbal emotional vocal utterances (e.g., laughing, shrieking, moaning) and emotional intonation patterns (e.g., scolding, lamenting, caressing) can be shown to have their acoustic and emotional counterparts in the vocal repertoire of the squirrel monkey. This makes the latter an attractive model for investigations on the neural control of human emotional vocal utterances. Neurophysiological investigations in the squirrel monkey suggest that the cerebral control of emotional vocal utterances is organized hierarchically (Fig. 3). The lowest level - above that of the motor neurons - is represented by the reticular formation of the lateral pons and medulla; here, the motor coordination of laryngeal, respiratory and articulatory movements takes place. The next level is represented by the periaqueductal grey and laterally bordering tegmentum of the caudal midbrain. This area serves to couple specific motivational states to their corresponding vocal expressions. It is a necessary relay station for all vocalization-inducing stimuli. The periaqueductal area receives its input partly from limbic motivation-controlling regions (amygdala, hypothalamus, midline thalamus), partly from sensory pathways (collaterals of spinothalamic tract, fibers from superior and inferior colliculus), and partly from the anterior cingulate cortex. The latter represents the highest level within the system and seems to be responsible for the volitional control of emotional vocal utterances.