Functional ultrasound imaging of the dynamical structure of socio-vocal network in monkeys
Daniel Y. Takahashi, Ahmed El Hady, Gabriel Montaldo, Alan Urban, Asif A. Ghazanfar
Vocal communication is the quintessential form of social interaction. Humans and other animals coordinate their social behaviors by producing and perceiving distinct vocalizations. Brain networks related to vocal communication include areas at the intersection of social behavior and vocal production-perception networks. Recent studies of primate vocal communication focused on lateral cortical regions, despite the fact that medial cortical and subcortical areas constitute the main vocal production and social behavior network (SBN). Hence, we aim to unravel the brain wide network underlying social communication focusing on the role played by medial cortical and subcortical areas. We use as our model the marmoset monkey, a highly vocal New World species. To image large-scale neural activity, we use functional ultrasound imaging which has a large spatial coverage and high spatio-temporal resolution. Furthermore, we built a stochastic dynamical systems model of vocal behavior that interacts with the marmoset in a closed-loop to fully control the vocal interaction and make quantitative predictions about brain dynamics during communication. We first show the existence of a medial brain system at the intersection of vocal production-perception and SBN; we call it the socio-vocal network (SVN). These areas differentially respond to affiliative vocalizations—contact, trillphee, and trill calls—produced in different contexts, exhibiting the highest and quickest response to contact calls. Given that the contact calls reflect the highest arousal state of the vocalizing animal, this is consistent with the hypothesis that SVN is related to the monitoring of others motivational state through vocalization. Second, through a closed-loop interaction between the computational model and a marmoset, together with large-scale functional imaging, we found that the marmoset anterior cingulate cortex (which is part of SVN) and the model’s “SVN" are entrained. These results demonstrate what the SVN encompasses and its roles in vocal communication.