Synapses are a fundamental unit of computation in the brain. Far from being passive connections between spiking neurons, synapses display striking short-term dynamics, undergo long-term changes in strength, and sculpt network-level processes in a complex manner. These synaptic dynamics, both in time and across space, may be a fundamental determinant of population-level computations and behavioral output of the brain, yet their role in neuromodulatory circuits is relatively under-explored. First, I developed and validated a set of likelihood-based inference tools to quantify the dynamics of synaptic ensemble composition throughout development. Second, I examined network computations in the serotonergic dorsal raphe nucleus through a dynamical lens, exploring the role of short-term synaptic dynamics at sparse recurrent connections, and of distinct long-range synaptic inputs, in shaping the output of spiking populations. 1. Simulation-based inference of synaptic ensembles. Functional features of synapses are typically inferred by sampling small ensembles of synapses, yet it is unclear if such subsamples exhibit biases. I developed a statistical framework to address this question, using it to demonstrate that common bulk electrical stimulation methods for characterizing the fraction of silent synapses exhibit high bias and variance, and using typical sample sizes, possess insufficient statistical power for accurate inference. I developed and validated a novel synthetic likelihood-based inference approach based on a simulator of the underlying experimental methodology. This new estimator, made available in an object-oriented Python toolbox, reduces bias and variance compared to previously reported methods, and provides a scalable method for examining synaptic dynamics throughout development. These tools were validated by targeted recording from hippocampal CA1 neurons in juvenile mice, where they reveal fundamental tradeoffs between release probability, number of synapses sampled, and statistical power. 2. Synaptic dynamics and population computations in the serotonin system. This part is comprised of two manuscripts. First, in the dorsal raphe nucleus, I uncovered slow, inhibitory recurrent interactions between serotonin neurons that are generated by local serotonin release. These connections were probabilistic, displayed striking short-term facilitation, gated the spiking output of serotonin neurons, and could be activated by long-range excitatory input from lateral habenula, representing threat signals. Targeted physiology and modeling revealed that these recurrent short-term facilitation features generated paradoxical excitation-driven inhibition in response to high-frequency habenula input. These facilitation rules additionally supported winner-take-all dynamics at the population level, providing a contrastive operation between functionally distinct serotonergic ensembles. Behaviorally, activating long-range lateral habenula input to dorsal raphe nucleus generated a transient, frequency-dependent suppression of reward anticipation consistent with these recurrent dynamics, without modulating the underlying reward association itself. These dynamics, we suggest, support sharp behavioral state transitions in changing environments. In a second manuscript, I explored the multiplexing of distinct long-range inputs in serotonergic circuits through spike synchrony. I demonstrated that a population of serotonergic neurons receives input from both lateral habenula and prefrontal cortex. These inputs produced similar subthreshold events, but prefrontal cortex triggered spikes with much higher latencies, supporting a population synchrony code for input identity. These input-specific spike timing patterns could be read out by simple linear decoders with high accuracy, suggesting they could be demultiplexed by downstream circuits receiving sparse innervation by serotonergic axons. We uncovered a novel intracellular calcium conductance in serotonergic neurons that altered the spectral characteristics of membrane voltage in a manner sufficient to generate long-latency, power law-distributed spike times, suggesting a simple dynamical origin for the production of synchronous or asynchronous spiking. This work indicates that serotonergic circuits can multiplex distinct informational streams through population spike synchrony mechanisms. Together, these investigations reveal that the dynamics of short-term facilitation and synaptic ensemble composition can act as the fundamental substrate for flexible computation by spiking networks across the brain.

This dissertation aims to shed light on the ways in which our affective responses and subjective beliefs shape our reasoning about social events and targets. The human ability to reason about other people’s minds, and the social world in which we live, has been central to the field of psychology. However, that ability to make sense of the social world does not exist in isolation. Each social perceiver has idiosyncratic beliefs and identities. Perceivers also affectively respond to events and people in the world around them. Historically, the processes underlying affective processing, social cognition, and formed beliefs, have been studied in isolation, leading to a gap in our knowledge about their interactions. We conducted a set of experiments combining fMRI and behavioral methods to address this gap. The experiments used naturalistic stimuli, which allow related processes to co-occur in an ecologically valid way. The results of the experiments are described in three chapters, following a general introduction (Chapter 1). In Chapter 2, we show that the mentalizing regions of the brain represent a continuous affective response to social targets, and demonstrate a link between that response and the impression perceivers formed of those targets. In Chapter 3, we demonstrate that when presented with conflicting accounts of the same events, the subsequent event representation in participants medial prefrontal cortex is in concordance with perceivers’ beliefs about the events. In Chapter 4, we describe a cross-disciplinary study, informed by political scientific theories about the roots of polarization. In this study, we challenged partisan’s political beliefs and identities. We found that affective responding brain regions showed an effect of partisan information processing for both ideological beliefs and identity challenges. In addition, using two functional localizer tasks, we identified two sets of regions with differing functional profile within the mentalizing network. One set of regions showed the effect of partisan information processing only when perceivers’ ideology was challenged, while the other set showed the effect only when perceivers’ identity was challenged. Taken together, the results from these three studies expand our understanding of the mentalizing regions by suggesting that they represent not only the mental states of others, but also an affective response towards them. This work also reinforces our understanding of the differences in level of abstraction of the representation between prefrontal and parietal mentalizing regions. Lastly, the finding of different yet consequential activation profiles within the mentalizing network opens the door for further inquiries into the functional organization and representations within its constituting regions.

As humans we are not stuck in an everlasting present. Instead, we can project ourselves into both our personal past and future. Remembering the past and simulating the future are strongly interrelated processes. They are both supported by largely the same brain regions including the rostral and ventral medial prefrontal cortex (mPFC) but also the hippocampus, the posterior cingulate cortex (PCC), as well as other regions in the parietal and temporal cortices. Interestingly, this core network for episodic simulation and episodic memory partially overlaps with a brain network for evaluation and value-based decision making. This is particularly the case for the mPFC. This part of the brain has been associated both with a large number of different cognitive functions ranging from the representation of memory schemas and self-referential processing to the representation of value and affect. As a consequence, a unifying account of mPFC functioning has remained elusive. The present thesis investigates the unique contribution of the mPFC to episodic simulation by highlighting its role in the representation of memory schemas and value. In a first functional MRI and pre-registered behavioral replication study, we demonstrate that the mPFC encodes representations of known people as well as of known locations from participants’ everyday life. We demonstrate that merely imagined encounters with liked vs. disliked people at these locations can change our attitude toward the locations. The magnitude of this simulation-induced attitude change was predicted by activation in the mPFC during the simulations. Specifically, locations simulated with liked people exhibited significantly larger increases in liking than those simulated with disliked people. In a second behavioral study, we examined the mechanisms of simulation-based learning more closely. To this end, participants also simulated encounters with neutral people at neutral locations. Using repeated behavioral assessments of participants’ memory representations, we reveal that simulations cause an integration of memory representations for jointly simulated people and locations. Moreover, compared to the neutral baseline condition we demonstrate a transfer of positive valence from liked and of negative valence from disliked people to their paired locations. We also provide evidence that simulations induce an affective experience that aligns with the valence of the person and that this experience can account for the observed attitude change toward the location. In a final fMRI study, we examine the structure of memory representations encoded in the mPFC. Specifically, we provide evidence for the hypothesis that the mPFC encodes schematic representations of our social and physical environment. We demonstrate that representations of individual exemplars of these environments (i.e., individual people and locations) are closely intertwined with a representation of their value. In sum, our findings show that we can learn from imagined experience much as we learn from actual past experience and that the mPFC plays a key role in simulation-based learning. The mPFC encodes information about our environment in value-weighted schematic representations. These representations can account for the overlap of mnemonic and evaluative functions in the mPFC and might play a key role in simulation-based learning. Our results are in line with a view that our memories of the past serve us in ways that are oriented toward the future. Our ability to simulate potential scenarios allows us to anticipate the future consequences of our choices and thereby fosters farsighted decision making. Thus, our findings help to better characterize the functional role of the mPFC in episodic future simulation and valuation.

The lateral prefrontal cortex (LPFC) and the anterior cingulate cortex (ACC) are two key regions of the frontal executive control network. Ascending cholinergic pathways differentially innervate these two functionally distinct cortices to modulate arousal and motivational signaling for higher-order functions. The action of acetylcholine (ACh) in sensory cortices is constrained by layer, anatomical cell type, and subcellular localization of distinct receptors, but little is known about the nature and organization of frontal-cholinergic circuitry in primates. In this dissertation, we characterized the anatomical localization of muscarinic acetylcholine receptors (mAChRs), m1 and m2–the predominant subtypes in the cortex–and their expression profiles on distinct cell types and pathways in ACC and LPFC of the rhesus monkey, using immunohistochemistry, anatomical tract-tracing, whole cell patch-clamp recordings, and single nucleus RNA sequencing. In the first series of studies (Chapter 2), we used immunohistochemistry and high-resolution confocal microscopy to reveal regional differences in m1 and m2 receptor localization on excitatory pyramidal and inhibitory neuron subpopulations and subcellular compartments in ACC (A24) versus LPFC (A46) of adult rhesus monkeys (Macaca mulatta; aged 7-11 yrs; 4 males and 2 females). The ACC exhibited a greater proportion of m2+ inhibitory neurons and a greater density of presynaptic m2+ receptors localized on inhibitory (VGAT+) terminations on pyramidal neurons compared to the LPFC. This result suggests a greater cholinergic suppression of GABAergic neurotransmission in ACC. In a second set of experiments (Chapter 3), we examined the heterogeneity of m1 and m2 laminar expression in functionally distinct ACC areas A24, A25, and A32. These differ in their connections with higher order cortical areas and limbic structures, such as the amygdala (AMY). The density of m1+ and/or m2 expressing (m1+/m2+) pyramidal neurons was significantly greater in A24 compared to A25 and to A32, while A25 exhibited a significantly greater density of m2+VGAT+ terminals. In addition, we examined the substrates for cholinergic modulation of long-range cortico-limbic processing using bidirectional neural tracers to label one specific subtype, the AMY-targeting projection neurons in these ACC areas. Compared to A24 and A32, the limbic ventral A25 had a greater density of m1+/m2+ AMY-targeting pyramidal neurons across upper layers 2-3 and deep layers 5-6, suggesting stronger cholinergic modulation of amygdalar outputs. Lastly (Chapter 4), we assessed the functional effects of cholinergic modulation on excitatory and inhibitory synaptic activity as well as the molecular signatures related to m1 and m2 receptor expression. In experiments using in vitro whole-cell patch-clamp recordings of layer 3 pyramidal neurons in ACC and LPFC, we found that application of the cholinergic agonist carbachol (CCh) significantly decreased the frequency of excitatory postsynaptic currents (EPSCs) to a greater extent in ACC A24 than in LPFC A46. Using single nucleus RNA sequencing, we found that enriched m1 and m2 transcriptional profiles in distinct cell-types and frontal areas (ACC A24 and LPFC A46) had differentially expressed genes associated with down-stream signaling cascades related to synaptic signaling and plasticity. Together, these data reveal the anatomical, functional, and transcriptomic neural substrates of diverse cholinergic modulation of local excitatory and inhibitory circuits and long-range cortico-limbic pathways in functionally-distinct ACC and LPFC frontal areas that are important for cognitive-emotional integration.

Amyotrophic lateral sclerosis and frontotemporal dementia (ALS-FTD) represents a spectrum of neurodegenerative disease with clinical presentations ranging from progressive paralysis to cognitive impairment. Approximately 15% of ALS-FTD patients initially presenting with motor symptoms also receive a diagnosis of dementia, but a majority of these patients demonstrate some level of cognitive impairment over the course of disease. Identifying molecular pathways that contribute to the development of cognitive deficits in ALS-FTD has thus far been limited by the quality of clinical information and postmortem tissue preservation as well as available technologies. This thesis aims to investigate early stages of cognitive involvement in ALS-FTD using postmortem tissues from a cohort of non-demented ALS patients who have had cognitive and pathological phenotyping. Spatially resolved transcriptome profiling of prefrontal cortex tissues from this cohort identifies dysregulated pathways in non-motor regions, contributing to our understanding of molecular perturbations underlying cognitive impairment in ALS-FTD.

Neural coding is essential for understanding how the activity of individual neurons or ensembles of neurons relates to cognitive processing of the world. Neurons can code for multiple variables simultaneously and neuroscientists are interested in classifying neurons based on the variables they represent. Building a model identification paradigm to identify neurons in terms of their coding properties is essential to understanding how the brain processes information. Statistical paradigms are capable of methodologically determining the factors influencing neural observations and assessing the quality of the resulting models to characterize and classify individual neurons. However, as neural recording technologies develop to produce data from massive populations, classical statistical methods often lack the computational efficiency required to handle such data. Machine learning (ML) approaches are known for enabling efficient large scale data analysis; however, they require huge training data sets, and model assessment and interpretation are more challenging than for classical statistical methods. To address these challenges, we develop an integrated framework, combining statistical modeling and machine learning approaches to identify the coding properties of neurons from large populations. In order to evaluate our approaches, we apply them to data from a population of neurons in rat hippocampus and prefrontal cortex (PFC), to characterize how spatial learning and memory processes are represented in these areas. The data consist of local field potentials (LFP) and spiking data simultaneously recorded from the CA1 region of hippocampus and the PFC of a male Long Evans rat performing a spatial alternation task on a W-shaped track. We have examined this data in three separate but related projects. In one project, we build an improved class of statistical models for neural activity by expanding a common set of basis functions to increase the statistical power of the resulting models. In the second project, we identify the individual neurons in hippocampus and PFC and classify them based on their coding properties by using statistical model identification methods. We found that a substantial proportion of hippocampus and PFC cells are spatially selective, with position and velocity coding, and rhythmic firing properties. These methods identified clear differences between hippocampal and prefrontal populations, and allowed us to classify the coding properties of the full population of neurons in these two regions. For the third project, we develop a supervised machine learning classifier based on convolutional neural networks (CNNs), which use classification results from statistical models and additional simulated data as ground truth signals for training. This integration of statistical and ML approaches allows for statistically principled and computationally efficient classification of the coding properties of general neural populations.

The Lateral Prefrontal Cortex (LPFC) and the Dorsal Premotor cortex (PMd) are two cortical structures that are involved in cognitive processes such as motor planning and decision-making. The LPFC is extensively connected to sensory, somatosensory, and motor cortices that help it control several cognitive functions [for review, see: (Tanji & Hoshi, 2008)]. Similarly, the PMd can integrate information from the prefrontal and motor cortex, acting as a link, in action planning and decision making [for review, see: (Hoshi & Tanji, 2007)]. Therefore, it is important to study the cortical pathways between these areas because of their common role in processing and selecting relevant information in tasks requiring decision-making. Using neural tract-tracing, immunolabeling and microscopy in rhesus monkeys (M. mulatta), we assessed the distribution and layer-specific organization of projection neurons from LPFC area 46 and PMd area 6 directed to the LPFC area 9. Our study revealed that projection neurons to area 9 were found originating from upper (L2-3) and deep (L5-6) layers of both areas, but with a slight upper layer bias. We found that the LPFC area 46 had a higher density of projection neurons directed to LPFC area 9 compared to the PMd area 6. Additionally, our data also revealed laminar differences in the perisomatic parvalbumin (PV) inhibitory inputs onto area 9 projection neurons, which were dependent on area of origin. Within ventral LPFC area 46, perisomatic PV+ inhibitory inputs onto upper layer projection neurons to area 9 was greater than those onto deep layer projection neurons. The opposite pattern was found for PMd area 6DR, where perisomatic PV+ inhibition onto deep layer projection neurons to area 9 was greater than those onto upper layer neurons. These findings provide additional insights into the layer-specific organization of prefrontal and premotor pathways that play an important role in action planning and decision-making.

Huntington Disease (HD) is a fatal neurodegenerative disorder that is characterized by motor, cognitive, and psychiatric symptoms. Although HD onset is determined by motor symptoms, psychiatric symptoms, like depression and aggression, can develop earlier, have a larger impact on quality of life, and are understudied due to stigma. Our lab has observed hyper aggression in our humanized HD mouse model (Hu97/18) compared to our knock-in HD mouse model (Q175FDN). We characterized these differences and found that the Hu97/18 mice overreact in neutral situations, behaving as if they are in threatening situations. We are now using this novel model of HD-related aggression to study its neurological basis. Increased reactive aggression has been linked to stress levels and the prefrontal cortex (PFC) due to its role in emotional regulation. This study seeks to determine if HD-related aggression is associated with increased stress levels and changes in the PFC. Our cortisol study shows that the Hu97/18 mice display significantly higher cortisol levels than baseline, suggesting a link between systemic stress and heightened aggression. Additionally, quantified PFC volumes show a moderate relationship between PFC volume and aggression in wild-type (WT) mice that is lost in the Hu97/18 mice. This data will help elucidate factors that modulate aggression in HD and may identify therapies with high potential to alleviate this devastating symptom in patients.

No description available.

No description available.