Student Theses and Dissertations

Date of Award

2024

Document Type

Thesis

Degree Name

Doctor of Philosophy (PhD)

Abstract

Learned vocal communication and spoken language are complex behaviors we have long sought to understand. Studying the neurobiology of speech and language will be advanced by investigating its component traits in model organisms. Most vertebrates share common brainstem circuits for vocalization, and they produce innate vocal repertoires over which the animals have little to no control. Vocal learners can imitate heard sounds, and thus have a high degree of control over their vocalizations. Vocal learning species have a direct projection from the pallial forebrain to the brainstem vocal motor neurons, which facilitate their vocal dexterity. Songbirds have been the standard model of vocal learning, but there are technical limits to our ability to test various hypotheses about the development and evolution of vocal learning circuits. Mice, which are more closely related to humans, have been found to exhibit some rudimentary features that are like those seen in vocal learning species. These include: 1) a direct, but sparse, projection from the primary motor cortex(M1) to the laryngeal motor neurons in the brainstem, termed the laryngeal motor cortex (LMC); 2) cortical lesions that damage this direct projecting population alter the frequency distribution of vocalizations; and 3) change their vocal syntax based on social context. Although mouse ultrasonic vocalizations (USVs) have received a great deal of attention and interest, the role these cortical circuits and the mechanisms by which they might affect USVs have been little explored. Here, we investigate the role of the motor cortex of mice in controlling vocal musculature, and we develop new methods that will allow us to gain more insights into the control mice have over their USVs. We performed intracortical microstimulation (ICMS) with paired electromyographic (EMG) recordings to test whether the direct projection previously identified in mice can generate laryngeal muscle contractions. Simultaneously, we recorded EMG signals from the anterior digastric, a jaw opening muscle, as a control. We found that the LMC population of neurons can generate laryngeal muscle contractions. We also identified the orofacial motor cortex (OMC) as another region of M1 that can generate laryngeal muscle contractions, although weaker than from LMC. The muscles responded with different latencies from the LMC and OMC stimulations suggestive of both indirect and direct brainstem motor neuron projections, respectively. Using a retrograde transsynaptic virus, we show that the region of M1 containing the LMC has neurons that represent the larynx as well as jaw and forelimb muscles; in a small proportion of neurons, two muscles were represented by single neurons. Using an anterograde transsynaptic tracer from the OMC, we found that there are direct projections to the motor neurons of the jaw. Lastly, chemical lesions of OMC led to a modest change in the number of USVs per sequence. To test the degree of control mice have over their USVs, we developed a suite of tools to train mice in an operant vocal task. To detect and classify USVs in real-time, we developed a software called Analysis of Mouse Vocal Communication (AMVOC). We combined AMVOC with other open-source hardware and software to design an operant training paradigm to that lets us test the volitional control of mouse vocal behavior. We provide a proof-of-principle application of this system to train mice to increase vocalizations for food reward. Mice had been assumed to lack a functional cortical representation of the larynx, and similar assumptions have been made about other vocal non-learners. In contrast, the results of my thesis provide evidence that M1 in a vocal non-learner can influence vocal musculature, consistent with the continuum hypothesis of vocal learning. We also demonstrate that the representations of muscles for different behaviors across mouse M1 are highly intermixed, sharing both cortical space and single neurons. These results offer new insights into the origin and evolution of laryngeal control by cortical circuits, suggesting that these circuits are both more commonly distributed across mammalian species and that they may have arisen earlier than was previously hypothesized. Further, the results presented here will allow us to better understand the shared features of vocal production that are integral to better understanding human speech.

Comments

A Thesis Presented to the Faculty of The Rockefeller University in Partial Fulfillment of the Requirements for the degree of Doctor of Philosophy

Download

Included in

Life Sciences Commons

Share

COinS