Browsing by Author "Mooney, Richard"

Many of the skills we value most as humans, such as speech and learning to play musical instruments, are learned in the absence of external reinforcement. However, the model systems most commonly used to study motor learning employ learning paradigms in which animals perform behaviors in response to external rewards or punishments. Here I use the zebra finch, an Australian songbird that can learn its song as a juvenile in the absence of external reinforcement as well as modify its song in response to external cues as an adult, to study the circuit mechanisms underlying both internally and externally reinforced forms of learning. Using a combination of intersectional genetic and microdialysis techniques, I show that a striatonigral pathway and its downstream effectors, namely D1-type dopamine receptors, are necessary for both internally reinforced juvenile learning and externally reinforced adult learning, as wells as for song modification in response to social cues or to deafening. In addition, I employ optogenetic stimulation during singing to demonstrate that this striatonigral projection is sufficient to drive learning. Interestingly, I find that neither the striatonigral pathway nor D1-type dopamine receptors are necessary for recovery of pitch after externally driven pitch learning. In all, I establish that a common mechanism underlies both internally and externally reinforced vocal learning.

The ability to learn and to modify complex vocal sequences requires extensive practice coupled with performance evaluation through auditory feedback. An efficient solution to the challenge of vocal learning, stemming from reinforcement learning theory, proposes that an “actor” learns correct vocal behavior through the instructive guidance of an auditory “critic.” However, the neural circuit mechanisms supporting performance evaluation and even how “actor” and “critic” circuits are instantiated in biological brains are fundamental mysteries. Here, I use a songbird model to dissociate “actor” and “critic” circuits and uncover biological mechanisms for vocal learning.

First, I employ closed-loop optogenetic methods in singing birds to identify two inputs to midbrain dopamine neurons that operate in an opponent fashion to guide vocal learning. Next, I employ electrophysiological methods to establish a microcircuit architecture underlying this opponent mechanism. Notably, I show that disrupting activity in these midbrain dopamine inputs precisely when auditory feedback is processed impairs learning, showing that they function as “critics.” Conversely, I show that disrupting activity in a downstream premotor region prior to vocal production prevents learning, consistent with an “actor” role. Taken together, these experiments dissociate discrete “actor” and “critic” circuits in the songbird’s brain and elucidate neural circuit and microcircuit mechanisms by which “actors” and “critics” working cooperatively enable vocal learning.

Children learn crucial skills like speech by imitating the behavior of skilled adults. Similarly, juvenile zebra finches learn to sing by learning to imitate adults. This song learning process enables laboratory study of juvenile imitative learning. But it also poses behavioral quantification challenges. Zebra finches produce hundreds of thousands of complex vocalizations during vocal development. These undergo learned changes with respect to acoustic features that are relevant to the animal but experimentally unknown \textit{a priori}. Recent developments in machine learning provide tools to reduce the dimensionality of complex behaviors, plausibly simplifying this inference challenge. These tools have not been validated on or applied to song learning problems.

Here, I validate the use of an autoencoder to extract copying-relevant features from zebra finch song. Then, I develop tools to quantify developmental song change with respect to extracted features. In particular, I generate forward models that quantify developmental changes in syllable acoustic distributions. I also develop a method to score syllable maturity on a rendition-by-rendition basis. Both these techniques reveal circadian behavioral patterns that differ between normally developing and untutored juveniles, suggesting that tutoring not only sets target song acoustics; it directly affects intrinsic features of practice behavior. Critically, these tools enable making concrete predictions from otherwise abstract song learning theories.

Mutations of the FOXP2 gene impair speech and language development in humans and shRNA-mediated suppression of the avian orthologue FoxP2 disrupts song learning in juvenile zebra finches. How diminished FoxP2 levels affect vocal control and alter the function of neural circuits important to learned vocalizations remains unclear. Using a combination of behavioral analysis, in vivo intracellular recordings in anaesthetized birds, pharmacology and extracellular recordings in singing birds, I addressed how FoxP2 knockdown in songbird striatum affects vocal control and signal propagation through circuits important for the control of learned vocalizations. In summary, I found that FoxP2 knockdown in the songbird striatum disrupts developmental and social modulation of song variability. Recordings in anaesthetized birds show that FoxP2 knockdown interferes with D1R-dependent modulation of activity propagation in a corticostriatal pathway important to song variability, an effect that may be partly attributable to reduced D1R and DARPP-32 protein levels. Furthermore, recordings in singing birds reveal that FoxP2 knockdown prevents social modulation of singing-related activity in this pathway. These findings show that reduced FoxP2 levels interfere with the dopaminergic modulation of vocal variability, which may impede song and speech development by disrupting reinforcement learning mechanisms.

While learning in response to extrinsic reinforcement is theorized to be driven by dopamine signals that encode the difference between expected and experienced rewards, skills that enable verbal or musical expression can be learned without extrinsic reinforcement. Instead, spontaneous execution of these skills is thought to be intrinsically reinforcing. Whether dopamine signals similarly guide learning of these intrinsically reinforced behaviors is unknown. Juvenile zebra finches are distinguished by their ability to copy the song of an adult tutor, a spontaneous, intrinsically reinforced process. Here, I use the zebra finch as a model system to study the neural mechanisms that operate within a song-specialized region of the basal ganglia (sBG) to enable this remarkable form of motor learning. Using in vivo microdialysis and computational methods to quantify juvenile song development, I first determined that dopamine (DA) signaling in the sBG is necessary for song learning. Using genetically encoded DA sensors and fiber photometry, I showed that DA dynamics in the sBG faithfully track the learned quality of juvenile song performance on a rendition-by-rendition basis. Consequently, my experiments provide compelling evidence that DA functions in the sBG as a reward prediction error-like signal to drive song learning, a process that evolves spontaneously and does not depend on extrinsic reward or punishment. Furthermore, I found that DA release in the sBG is driven not only by inputs from midbrain DA neurons classically associated with reinforcement learning but also by song premotor “cortical” inputs, which act via local cholinergic signaling in the sBG to elevate DA during singing. While I was able to show that both cholinergic and dopaminergic signaling in the sBG are necessary for song learning, I further found that only DA tracks the learned quality of song performance. Therefore, dopamine dynamics in the basal ganglia encode performance quality to drive self-directed and long-term learning of natural behaviors.

The late phase of long-term synaptic potentiation (LTP) at glutamatergic synapses, which is thought to underlie the long lasting memory (at least hours), requires gene transcription in the nucleus. However, it remains elusive how signaling initiated at synapses during induction of LTP is transmitted into the nucleus to commence transcription. Using a combination of two-photon glutamate uncaging and a genetically encoded FRET sensor, I found that induction of synapse-specific LTP at only a few (3-7) dendritic spines leads to pronounced activation of extracellular signal-regulated kinase (ERK) in the nucleus and downstream phosphorylation of transcription factors, cAMP-response element-binding protein (CREB) and E26-like protein-1 (Elk-1). The underlying molecular mechanism of this nuclear ERK activation was investigated: it seems to involve activation of NMDA receptors, metabotrophic glutamate receptors, and the classical Ras pathway. I also found that the spatial pattern of synaptic stimulation matters: spatially dispersed stimulation over multiple dendritic branches activated nuclear ERK much more efficiently than clustered stimulation within a single dendritic branch. In sum, these results suggest that biochemical signals could be transmitted from individual spines to the nucleus following LTP induction and that such synapse-to-nucleus signaling requires integration across multiple dendritic branches.

Perceiving or producing complex vocalizations such as speech and birdsongs require the coordinated activity of neuronal populations, and these activity patterns can vary over space and time. How learned communication signals are represented by populations of sensorimotor neurons essential to vocal perception and production remains poorly understood. Using a combination of two-photon calcium imaging, intracellular electrophysiological recording and retrograde tracing methods in anesthetized adult male zebra finches (Taeniopygia guttata), I addressed how the bird's own song and its component syllables are represented by the spatiotemporal patterns of activity of two spatially intermingled populations of projection neurons (PNs) in HVC, a sensorimotor area required for song perception and production. These experiments revealed that neighboring PNs can respond at markedly different times to song playback and that different syllables activate spatially intermingled HVC PNs within a small region. Moreover, noise correlation analysis reveals enhanced functional connectivity between PNs that respond most strongly to the same syllable and also provides evidence of a spatial gradient of functional connectivity specific to PNs that project to song motor nucleus (i.e. HVC_RA cells). These findings support a model in which syllabic and temporal features of song are represented by spatially intermingled PNs functionally organized into cell- and syllable-type networks.

To successfully navigate the world, our sensory systems must process stimuli accurately during both rest and movement. Indeed, movements have been shown to modulate sensory systems at different levels. In audition, studies in humans and other animals have shown that movements strongly suppress auditory cortical responses to acoustic stimuli relative to rest. A largely untested idea is that this cortical suppression works to suppress responses to predictable acoustic consequences of movements, while enhancing sensitivity to novel stimuli. How this cortical suppression influences auditory perception and whether this suppression functions predictively as widely theorized remain unknown. Here, I trained head-fixed mice to lick in response to tones of different intensities during rest or running on a quiet treadmill. I observed that auditory detection was impaired during running compared to rest. Inactivating the auditory cortex impaired detection, and optogenetically activating secondary motor cortical axons in the auditory cortex during rest degraded detection similar to movement. Finally, movement-related impairment of auditory detection was specific to expected sounds following predictable sensorimotor experience. Overall, these findings support the idea that movement-related modulation of auditory cortical activity is behaviorally adaptive, selectively suppressing predictable movement-related sounds while enhancing sensitivity to novel stimuli.

Skilled movements are typically more variable during practice, promoting exploration, yet highly stereotyped during performance, favoring exploitation. How neurons encode and dynamically regulate motor variability across practice and performance states remains unknown. Songbirds sing more variable songs when practicing alone and highly stereotyped songs when performing to a female, providing a powerful system to explore how neural ensembles regulate motor variability. Here, I used this system to identify neural mechanisms underlying practice and performance. First, I used deep brain imaging techniques to demonstrate that spiny neurons (SNs) in the basal ganglia (BG) encode vocal variability during solo practice, and that SN activity is strongly suppressed to enable stereotyped song performance towards a female. Second, I showed that optogenetically inhibiting SNs reduces pitch variability to female-directed levels. Third, I collaborated with Dr. John Pearson’s lab to uncover a coding scheme whereby specific patterns of SN activity map onto distinct spectral variants of syllables during vocal practice. Lastly, I use photometry, anatomical tracing, molecular profiling, and ex vivo physiology to establish that adrenergic signaling in the BG regulates vocal variability by directly suppressing SN activity. I conclude that SN ensembles encode and drive vocal exploration during practice, and the social context-dependent noradrenergic regulation of SN activity enables stereotyped and highly precise vocal performance.

Vocal communication is a key behavior by which mammals form social bonds and convey information about social status and mating fitness. However, the cellular and synaptic nature of the neural circuits that adaptively regulate vocalization as a function of the individual’s social and environmental context remain unknown. Here, I helped pioneer the use of an intersectional genetic method to identify a subpopulation of neurons in the midbrain periaqueductal gray (PAG) of the mouse that act on downstream vocal-patterning circuits to gate ultrasonic courtship vocalizations. Next, I used transsynaptic tracing to identify two populations of inhibitory neurons that lie upstream of these PAG-USV neurons that exert opposing effects on USV production. I used molecular profiling, optogenetics, and circuit dissection in brain slices to establish that PAG-projecting GABAergic neurons in the preoptic hypothalamus promote USV production. In contrast, I found that PAG-projecting GABAergic neurons in the central-medial boundary zone of the amygdala suppress USV production without disrupting non-vocal social behavior. Finally, I used fiber photometry during free behaviors to reveal that affiliative social and sexual interactions excite USV-promoting preoptic neurons while innately aversive stimuli activate USV-suppressing amygdala neurons. These experiments provide an important step forward in mapping the brain-wide networks that regulate vocalizations as a function of social and environmental contexts.

The mouth, throat, and face contain numerous muscles that participate in a large variety of orofacial behaviors. The jaw and tongue can move independently, and thus require a high degree of coordination among the muscles that move them to prevent self-injury. However, different orofacial behaviors require distinct patterns of coordination between these muscles. The method through which motor control circuitry might coordinate this activity has yet to be determined. Electrophysiological, immunohistochemical, and retrograde tracing studies have attempted to identify populations of premotor neurons which directly send information to orofacial motoneurons in an effort to identify sources of coordination. Yet these studies have not provided a complete picture of the population of neurons which monosynaptically connect to jaw and tongue motoneurons. Additionally, while many of these studies have suggested that premotor neurons projecting to multiple motor pools may play a role in coordination of orofacial muscles, no clear functional roles for these neurons in the coordination of natural orofacial movements has been identified.

In this dissertation, I took advantage of the recently developed monosynaptic rabies virus to trace the premotor circuits for the jaw-closing masseter muscle and tongue-protruding genioglossus muscle in the neonatal mouse, uncovering novel premotor inputs in the brainstem. Furthermore, these studies identified a set of neurons which form boutons onto motor neurons in multiple motor pools, providing a premotor substrate for orofacial coordination. I then combined a retrogradely traveling lentivirus with a split-intein mediated split-Cre recombinase system to isolate and manipulate a population of neurons which project to both left and right jaw-closing motor nuclei. I found that these bilaterally projecting neurons also innervate multiple other orofacial motor nuclei, premotor regions, and midbrain regions implicated in motor control. I anatomically and physiologically characterized these neurons and used optogenetic and chemicogenetic approaches to assess their role in natural jaw-closing behavior, specifically with reference to bilateral masseter muscle electromyogram (EMG) activity. These studies identified a population of bilaterally projecting neurons in the supratrigeminal nucleus as essential for maintenance of an appropriate level of masseter activation during natural chewing behavior in the freely moving mouse. Moreover, these studies uncovered two distinct roles of supratrigeminal bilaterally projecting neurons in bilaterally synchronized activation of masseter muscles, and active balancing of bilateral masseter muscle tone against an excitatory input. Together, these studies identify neurons which project to multiple motor nuclei as a mechanism by which the brain coordinates orofacial muscles during natural behavior.

Auditory sensations can arise from objects in our environment or from our own actions, such as when we speak or make music. We must able to distinguish such sources of sounds, as well as form new associations between our actions and the sounds they produce. The brain is thought to accomplish this by conveying copies of the motor command, termed corollary discharge signals, to auditory processing brain regions, where they can suppress the auditory consequences of our own actions. Despite the importance of such transformations in health and disease, little is known about the mechanisms underlying corollary discharge in the mammalian auditory system. Using a range of techniques to identify, monitor, and manipulate neuronal circuits, I characterized a synaptic and circuit basis for corollary discharge in the mouse auditory cortex. The major contribution of my studies was to identify and characterize a long-range projection from motor cortex that is responsible for suppressing auditory cortical output during movements by activating local inhibitory interneurons. I used similar techniques to understand how this circuit is embedded within a broader neuromodulatory brain network important for learning and plasticity. These findings characterize the synaptic and circuit mechanisms underlying corollary discharge in mammalian auditory cortex, as well as uncover a broad network interaction potentially used to pattern neural associations between our actions and the sounds they produce.

Ventral tegmental area dopamine (VTADA) neurons fire in a manner consistent with Reward Prediction Error, with better-than-expected and worse-than-expected outcomes correlating with bursts and pauses, respectively. Burst and pause firing dynamics are believed to be responsible for driving associative learning, yet interrogating this causality, and understanding how these firing patterns are synaptically created within endogenous neural circuits, has been technically difficult. Utilizing a novel tool, DART (drug acutely restricted by tethering), paired with a multiplexed cue-reward associative learning task and in vivo neural recordings, I explore which classes of endogenous synaptic inputs to VTADA neurons create their canonical firing dynamics, and their role in the associated reward learning behaviors. My key finding is that antagonizing GABAA receptors on VTADA neurons decreases the pauses in firing these cells exhibit, but also accelerates extinction learning in response to unexpected reward omission. In the same mice, the manipulation had no impact on conditioning to a novel cue-reward pairing, indicating that positive-valence learning was unperturbed. This dissertation work provides critical insight into the neural circuitry underlying adaptive behaviors by creating a new framework for understanding conditioning and extinction as anti-correlated behaviors, and by establishing a novel role for direct inhibitory GABAA signaling to VTADA cells in conditioned conviction.

It has been suggested that primate brains are inherently biased towards gathering and processing the social information present in the world. In fact, the neural network that mediates our engagement with the external world - the default mode network (DMN) ¬- is strongly convergent with the neural circuitry for social cognition. The posterior cingulate (PCC) is believed to be a key node in both the DMN and in social cognition. Human and non-human primate studies have demonstrated a role for the PCC in outcome monitoring: it tracks rewards, subjective values of choices, task engagement and global choice strategies. It is also implicated in social cognition. Human studies show that PCC activity varies with the recall of autobiographical memories and exposure to social stimuli. While several electrophysiological studies explicate the response of PCC neurons to non-social outcome monitoring and valuation, there is a lack of similar studies for social valuation. This thesis is concerned with characterizing the neuronal responses in the PCC to social stimuli and determining whether social valuation occurs in the PCC in a manner similar to that previously described for non-social outcomes. I recorded the single unit activity of neurons in the PCC of rhesus macaques while they performed behavioral tasks that required attending to the faces of high-status or low-status individuals. Monkeys valued the faces of high-status individuals more than low-status individuals, though they were equally likely to identity and overtly attend to faces of both social classes. This differential valuation of face stimuli was represented in the firing activity of PCC neurons, with higher neuronal activity seen in response to subordinate faces as compared to dominant ones. Cells in the PCC did not track the individual identity of the presented faces. Furthermore, neuronal activity in the PCC predominantly tracked social value, and not non-social reward delivery as previously reported. Neuronal activity also predicted task engagement, with higher firing rates being predictive of a decrease in task engagement. To summarize, the PCC is biased towards social information processing, and neuronal activity in the PCC tracks social category information and the level of task engagement.

The ability of young children to vocally imitate the speech of adults is critical for speech learning. Vocal imitation requires exposure to an external auditory model and the use of auditory feedback to adaptively modify vocal output to match the model. Despite the importance of vocal imitation to human communication and social behavior, it remains unclear how these two types of sensory experience, model exposure and feedback, act on sensorimotor networks controlling the learning and production of learned vocalizations. Using a combination of longitudinal in vivo imaging of neuronal structure and electrophysiological measurements of neuronal function, I addressed the questions of where, when, and how these two types of sensory experience act on sensorimotor neurons important to singing and song learning in zebra finches. The major finding of these experiments is that synaptic inputs onto neurons in HVC, a sensorimotor nucleus important to singing and song learning, are sensitive to tutor song experience and deafening. Thus, these findings for the first time link auditory experiences important to vocal imitation to synaptic reorganization in sensorimotor neurons important to behavior.