Cortical basis for skilled vocalization

Marmosets display remarkable vocal motor abilities. Macaques do not. What is it about the marmoset brain that enables its skill in the vocal domain? We examined the cortical control of a laryngeal muscle that is essential for vocalization in both species. We found that, in both monkeys, multiple premotor areas in the frontal lobe along with the primary motor cortex (M1) are major sources of disynaptic drive to laryngeal motoneurons. Two of the premotor areas, ventral area 6 (area 6V) and the supplementary motor area (SMA), are a substantially larger source of descending output in marmosets. We propose that the enhanced vocal motor skills of marmosets are due, in part, to the expansion of descending output from these premotor areas.

Speech is a uniquely human form of communication which uses vocalization to express thoughts and feelings. Vocalization is built on the exquisitely coordinated control over respiration, phonation, and articulation. Historically, the enhanced vocal motor skills of humans have been attributed to alterations in the peripheral mechanisms for sound production (1, 2). However, recent studies of laryngeal biomechanics have ruled out this explanation (3). Instead, modifications in central neural circuits are the likely basis of the enhanced vocal abilities of humans (1). Here, we used a comparative approach to identify the adaptations in the cerebral cortex that provide a substrate for the enhanced vocal motor abilities of some monkeys.Our experiments compared the areas of the cerebral cortex that are involved in the control of a laryngeal muscle in macaques and marmosets. We selected these two monkey species because of the striking differences in their vocal behavior. Macaque vocalization is generally limited to spontaneous utterances of acoustically simple calls which relate the animal’s emotional and motivational state (4). In the laboratory setting, it is difficult for researchers to elicit macaque vocalizations and for the monkeys to suppress spontaneous calls (5, 6). In contrast, marmosets readily vocalize in the laboratory setting. These monkeys naturally exhibit vocal turn taking with multiple back-and-forth exchanges that entrain to each other just as in human conversation (7–10). Marmosets can modulate the amplitude (11), timing (9, 11), and pitch (12) of their calls to compensate not only for physical noise but also for physical distance between conspecifics. Overall, marmosets demonstrate vocal skills and experience-dependent vocal production not observed in macaques (13–15).To identify areas of the cerebral cortex that are involved in vocalization, we used retrograde transneuronal transport of rabies virus from the cricothyroid muscle. We selected the cricothyroid because it is the laryngeal muscle that is most specifically related to vocal motor control. The cricothyroid is an intrinsic laryngeal muscle that when active increases tension on the vocal folds (4). This muscle is unique in controlling vocal pitch while contributing little to other laryngeal functions, such as swallowing and airway regulation (4).