Perceptual decision making is a cognitive process that involves transforming sensory evidence into a decision and behavioral response through accumulating sensory information over time. Previous research has identified some temporally distinct components during the decision process; however, not all aspects of a perceptual decision are characterized by the post-stimulus activity. Using single-trial analysis with temporal localization techniques, we are able to identify a cascade of cognitive events associated with perceptual decision making, including what happens outside the period of evidence accumulation. The goal of this dissertation is to elucidate the association between neural correlates of these cognitive events. We design a set of experimental paradigms based on visual discrimination of scrambled face, car and house images and analyze EEG evoked potentials and oscillations using advanced machine learning and statistical analysis approaches. We first exploit the correlation between pre-stimulus attention and oscillatory activity and investigate such covariation within the context of behaviorally-latent fluctuations in task-relevant post-stimulus neural activity. We find that early perceptual representations, rather than temporally later neural correlates of the perceptual decision, are modulated by pre-stimulus brain state. Secondly, we demonstrate that the visual salience of stimulus image, being a surrogate for the decision difficulty, differentially modulates exogenous and endogenous oscillations at different times during decision making. This may reflect underlying information processing flow and allocation of attentional resources during the visual discrimination task. Finally, to study the effect of visual salience and value information of stimulus on feedback processing, we propose a model that can estimate expected reward and reward prediction error on a single-trial basis by integrating value information with perceptual decision evidence characterized by single-trial decoding of EEG. Taken together, these results provide a complete temporal characterization of perceptual decision making that includes the pre-stimulus brain state, the evidence accumulation during decisions and the post-feedback response evaluation.

Neurotransmitter:sodium symporter (NSS) proteins, the targets of antidepressants and psychostimulants, clear neurotransmitters from the synaptic cleft in a Na+-coupled transport mechanism. Transport is thought to occur via conformational rearrangements that alternately expose the substrate-binding site to each side of the membrane, but little is known about the mechanism by which ligand binding coordinates motions at the two faces. In this dissertation, single-molecule fluorescence resonance energy transfer (smFRET) techniques are used to image the dynamics of the prokaryotic NSS LeuT with sufficient resolution to describe the conformational states at both the intra- and extracellular faces for the first time. We found that the two sides do not move as a rigid body, contrary to popular models, and that previously undetected intermediate states are associated with transport activity. We also describe how ions and substrates influence conformational dynamics to create a productive transport cycle.

The question of how motivation affects our behavior is a long debated issue. Beyond pleasure and pain, motivation is closely related to cognitive functions and a key player in the self-regulation. The relationship between cognition and motivation was investigated from several angles, but a parsimonious explanation still awaits. In order to create a framework to understand the interaction between cognition and motivation, I chose two aspects of this relationship. Executive functions are one of the most studied psychological concepts and their components closely resemble the units of motivational processes. Secondly, a specific neural signature, dopamine, was selected due to its involvement in both executive functions and motivational processes. To enable dissection of motivation, in this thesis, we used a translational and a multilevel approach. In the first part, we focused on schizophrenia, which has a clinical presentation of cognitive (especially executive functions) and motivational deficits. Using a transgenic animal model mimicking the dopaminergic dysfunction related to schizophrenia, we manipulated motivation genetically, behaviorally, and pharmacologically and presented the changes in interval timing function. Part 2 of this thesis consists of 3 studies performed in humans to delineate the role of motivational orientations as measured by regulatory focus and regulatory mode surveys. A probabilistic reversal task and an n-back task were used to explore different components of executive functions; namely maintenance and monitoring, updating the representations, switching, and behavioral inhibition. The results of these studies showed that specific motivational orientations and their interactions could predict cognitive performance.

Historically, there have been two lines of research on mammalian hippocampus. The first one is concerned with the role of hippocampus in formations of new memories and owes its origin to the seminal study by Brenda Milner and William Scoville of a single memory disorder patient, widely known as H.M. The second line of research views the hippocampus as the brain area concerned with orienting and navigating in space. It started with John O’Keefe’s discovery of place cells, pyramidal neurons in the CA3 area of hippocampus, that fire when the animal enters a particular place in its environment. I argue that both lines of discoveries seem to be consistent with a more general view of hippocampus as a brain area strongly involved in the integration of sensory, and possibly internal, information. The first part of the thesis presents an investigation of the effect of limited connectivity constraint on the model network in the framework of pattern classification. It is shown that feed-forward neural classifiers with numerous long range connections can be replaced by networks with sparse feed-forward connectivity and local recurrent connectivity without sacrificing the classification performance. The limited connectivity constraint is relevant for most biological networks, and especially for the hippocampus. The second part describes a decoding analysis from the calcium signal recorded in mouse dentate gyrus. The animal’s position can be decoded with approximately 10cm accuracy and the neural representation of position in the dentate gyrus have close to maximal dimensionality. The analysis also suggests that cells with single firing field and cells with multiple firing fields contribute approximately equal amount of information to the decoder.

Within the past thirty years, the discovery and characterization of the microtubule-associated motor proteins, kinesins and cytoplasmic dynein, has radically expanded our understanding of intracellular trafficking and motile phenomena. Nevertheless, the mechanisms by which eukaryotic cells integrate motor functionality and cargo interactions over multiple subcellular domains in a spatiotemporally controlled way remain largely mysterious. During transport within the neuronal axon, dynein and the kinesins run in opposite directions along uniformly polarized microtubule tracks, so that each motor must switch between active transport and being, itself, a cargo in order to be properly positioned and carry out its function. The axon thus represents a model system in which to study the regulatory mechanisms governing intracellular transport, especially under conditions when it must be modulated in response to changing environmental cues, such as during axon outgrowth and development. Recently, the localization of certain messenger RNAs and their local translation to yield protein has emerged as a critical process for the development of axons and other neuronal compartments. I observed that transcripts encoding the dynein cofactors Lis1 and dynactin are among those localized to axons, so I hypothesized that stimulus-dependent changes in axonal transport may occur via local synthesis of dynein cofactors. In these studies, I have shown that different conditions of nerve growth factor signaling on developing axons trigger acute changes in the transport of various axonal cargoes, contemporaneous with rapid translational activation and production of Lis1 and dynactin’s main subunit, p150Glued, within the axons themselves. Differential synthesis of these cofactors in axons was confirmed to be required for the observed stimulus-dependent transport changes, which were completely prevented by axon-specific pharmacologic inhibition of protein synthesis or RNA interference targeted against Lis1 and p150Glued. In fact, Lis1 was, in an apparent paradox, locally synthesized in response to both nerve growth factor stimulation and withdrawal. I demonstrated that this is due to the fact that Lis1 is produced from a heterogeneous population of localized transcripts, differentiated chiefly by whether they interact with the RNA-binding protein APC. Preventing the binding of APC to Lis1 transcripts thus inhibited axonal synthesis of Lis1 and its resultant transport effects under conditions of nerve growth factor stimulation, while having no bearing on the similar phenomena seen during nerve growth factor withdrawal. This demonstrates that association with RNA-binding proteins can functionally distinguish sub-populations of localized messenger RNAs, which, in turn, provides a foundation for mechanistically understanding how localized protein synthesis is coupled to specific stimuli. Axonally synthesized Lis1 also was shown to have a particular role in mediating transport of a retrograde death signal originating in nerve growth factor-deprived axons, as neurons exhibited greatly reduced cell death when axonal synthesis of Lis1 was blocked. Through the application of pharmacologic agents inhibiting different steps in the propagation of this pro-apoptotic signal, I established that the signal depends upon effective endocytosis and the activity of glycogen synthase kinase 3β. It is therefore likely that the retrogradely transported signaling cargo in question is a glycogen synthase kinase 3β-containing endosome or multivesicular body—a type of large cargo consistent with Lis1’s known role in adapting the dynein motor for high-load transport. Preliminary results further indicate that axons exposed to another type of degenerative stress, in the form of toxic amyloid-β oligomers, may also employ local synthesis of Lis1 as a means of regulating transport and survival signaling. These findings establish a previously undescribed mechanism of regulating dynein activity and cargo interactions through local synthesis of its cofactors, allowing for rapid responses to environmental cues and stimuli that are especially relevant during the development of the nervous system. In addition to illustrating a regulatory principle that may be generally applicable to subcellular compartments throughout polarized cells, these studies provide new insights into intracellular transport disruptions that occur in lissencephaly, neurodegeneration, and other human disease states.

The hippocampus is traditionally thought to transmit contextual information to limbic structures where it acquires valence. Using freely moving calcium imaging and optogenetics, we show that while the dorsal CA1 subregion of the hippocampus is enriched in place cells, ventral CA1 (vCA1) is enriched in anxiety cells that are both activated by anxiogenic environments and required for avoidance behavior. Imaging cells defined by their projection target revealed that anxiety cells were enriched in the vCA1 population projecting to the lateral hypothalamic area (LHA), but not to the basal amygdala (BA). Consistent with this selectivity, optogenetic activation of vCA1 terminals in LHA, but not BA increased anxiety and avoidance, while activation of terminals in BA, but not LHA impaired contextual fear memory. Thus, the hippocampus encodes not only neutral but also valence-related contextual information, and the vCA1-LHA pathway is a direct route by which the hippocampus can influence innate anxiety behavior.

Neurons are constantly integrating information from external and internal sources, causing them to spike at particular times. The exact timing of spikes is determined by a neuron's intrinsic properties, as well as the interplay between local excitatory and inhibitory inputs. Although inhibitory interneurons have been extensively studied, their contribution to neuronal integration and spike timing remains poorly understood. To elucidate the functional role of GABAergic interneurons during cortical activity, we combined molecular identification of interneurons, two photon imaging and electrophysiological recordings in mouse thalamocortical slices. In this preparation, cortical UP states, a network state characterized by prolonged periods of depolarization and synchronized spiking, can be evoked by thalamic stimulation and can also occur spontaneously. To assay the role of inhibition, we first characterized the firing properties of Parvalbumin (PV) and Somatostatin (SOM) interneurons during UP states activity, and found a higher probability and rate of spiking in these two subtypes compared to excitatory cells. These subtypes did not display differential timing of activation during the evoked response. Furthermore, calcium imaging showed low correlations among PV and SOM interneurons, indicating that neurons sharing these neurochemical markers do not coordinate their firing. Intracellular recordings confirmed that nearby interneurons, known to be electrically coupled, do not display more synchronous spiking than excitatory cells, suggesting that this coupling may not function to synchronize the activity of interneurons on fast time scales¬¬¬. After characterizing inhibitory interneuron outputs, we next studied the timing and correlation of inhibitory inputs, which we isolated from excitatory inputs by voltage clamping at the reversal for excitation (0mV) or inhibition (-70mV). In both thalamically triggered and spontaneous activations, IPSCs between cell pairs were remarkably well correlated, with correlation coefficients reaching over .9 in some cases. This high degree of correlation has previously been assumed to be due to interneuron synchrony, but our population imaging and paired recordings did not support this view. In addition, we found that the connection rate between interneurons is very high (~80%), and quantal analysis revealed that each IPSC recorded in neighboring cells during an UP state could be due to a single presynaptic interneuron. Therefore, we explain the high IPSCs correlations in nearby pyramidal cells are emerging from the common input from individual interneurons, rather than from synchronization of interneuron activity across the population. In a final set of experiments, we found that a partial pharmacological block of inhibitory signaling increased EPSC correlations. Our data support a model in which inhibitory neurons do not fire in a correlated fashion but have strong, dense connections to pyramidal neurons that serve to prevent local excitatory synchrony during UP states. This would mean that inhibition may not, as previously thought, serve to synchronize the firing of excitatory cells, but have precisely the opposite effect, decorrelating their activity by breaking down their coordinated firing. This is consistent with the hypothesis that pyramidal cells are carrying out an essentially integrative function in the circuit and that interneurons expand the temporal dynamic range of this integration.

A key role in simplified models of neural circuitry (Wilson and Cowan, 1972) is played by the matrix of synaptic weights, also called connectivity matrix, whose elements describe the amount of influence the firing of one neuron has on another. Biologically, this matrix evolves over time whether or not sensory inputs are present, and symmetries possessed by the internal dynamics of the network may break up spontaneously, as found in the development of the visual cortex (Hubel and Wiesel, 1977). In this thesis, a full analytical treatment is provided for the simplest case of such a biological phenomenon, a single dendritic arbor driven by correlation-based dynamics (Linsker, 1988). Borrowing methods from the theory of Schrödinger operators, a complete study of the model is performed, analytically describing the break-up of rotational symmetry that leads to the functional specialization of cells. The structure of the eigenfunctions is calculated, lower bounds are derived on the critical value of the correlation length, and explicit expressions are obtained for the long-term profile of receptive fields, i.e. the dependence of neural activity on external inputs. The emergence of a functional architecture of orientation preferences in the cortex is another crucial feature of visual information processing. This is discussed through a model consisting of large neural layers connected by an infinite number of Hebb-evolved arbors. Ohshiro and Weiliky (2006), in their study of developing ferrets, found correlation profiles of neural activity in contradiction with previous theories of the phenomenon (Miller, 1994; Wimbauer, 1998). The theory proposed herein, based upon the type of correlations they measured, leads to the emergence of three different symmetry classes. The contours of a highly structured phase diagram are traced analytically, and observables concerning the various phases are estimated in every phase by means of perturbative, asymptotic and variational methods. The proper modeling of axonal competition proves to be key to reproducing basic phenomenological features. While these models describe the long-term effect of synaptic plasticity, plasticity itself makes the connectivity matrix highly dependent on particular histories, hence its stochasticity cannot be considered perturbatively. The problem is tackled by carrying out a detailed treatment of the spectral properties of synaptic-weight matrices with an arbitrary distribution of disorder. Results include a proof of the asymptotic compactness of random spectra, calculations of the shape of supports and of the density profiles, a fresh analysis of the problem of spectral outliers, a study of the link between eigenvalue density and the pseudospectrum of the mean connectivity, and applications of these general results to a variety of biologically relevant examples. The strong non-normality of synaptic-weight matrices (biologically engineered through Dale’s law) is believed to play important functional roles in cortical operations (Murphy and Miller, 2009; Goldman, 2009). Accordingly, a comprehensive study is dedicated to its effect on the transient dynamics of large disordered networks. This is done by adapting standard field-theoretical methods (such as the summation of ladder diagrams) to the non-Hermitian case. Transient amplification of activity can be measured from the average norm squared; this is calculated explicitly for a number of cases, showing that transients are typically amplified by disorder. Explicit expressions for the power spectrum of response are derived and applied to a number of biologically plausible networks, yielding insights into the interplay between disorder and non-normality. The fluctuations of the covariance of noisy neural activity are also briefly discussed. Recent optogenetic measurements have raised questions on the link between synaptic structure and the response of disordered networks to targeted perturbations. Answering to these developments, formulae are derived that establish the operational regime of networks through their response to specific perturbations, and a minimal threshold is found to exist for counterintuitive responses of an inhibitory-stabilized circuit such as have been reported in Ozeki et al. (2016), Adesnik (2016), Kato et al. (2017). Experimental advances are also bringing to light unsuspected differences between various neuron types, which appear to translate into different roles in network function (Pfeffer et al., 2013; Tremblay et al., 2016). Accordingly, the last part of the thesis focuses on networks with an arbitrary number of neuronal types, and predictions are provided for the response of networks with a multipopulation structure to targeted input perturbations.

Human brain evolution granted us cognitive and behavioral capabilities that are unique amongst animals. SRGAP2 is a gene that was specifically duplicated in the human lineage and plays roles in the regulation of cortical development and synapse dynamics. As paralogs of one of the few known genes that regulates excitatory and inhibitory synapses concurrently, the duplications of SRGAP2 were well-positioned during human evolution to gain novel functions leading to the cognitive and behavioral phenotypes exhibited in humans. SRGAP2C, a human-specific paralog of the ancestral SRGAP2 gene, inhibits every known function of SRGAP2 and induces a phenotype similar to SRGAP2 knockdown. This induces neoteny in the maturation of synapses in mice, allowing us to study a putatively “human-like” phenotype in the mouse brain. While studies have been conducted on the effects of SRGAP2 manipulation in juvenile and young adult mice, its effects on older mice has yet to be determined. In this dissertation, we perform longitudinal imaging experiments to determine the effects of SRGAP2 manipulation in the cortex of adult mice. In Chapter 3, we first examine the effects of SRGAP2 knockdown on the spine dynamics on apical dendrites of layer 5 pyramidal cells in the barrel cortex of adult mice, determining how it regulates spine density, turnover, and survival at baseline and in response to sensory deprivation. In Chapter 4, we study how SRGAP2 knockdown affects the clustered formations of new dendritic spines on the apical dendrites of layer 5 pyramidal cells in the barrel cortex of adult mice. Together, these results represent the first demonstration of SRGAP2 regulating on synapse dynamics in vivo and show that SRGAP2 knockdown can be used to model human brain evolution in adult mice.

This thesis focuses on my work using optical techniques to study different brain regions in vivo. The ability to optically study neurons and the circuits they comprise in vivo is an important method to better understand their role in the healthy brain and their dysfunction in disease. The first part of my thesis focuses on my work using on a collaborative project using a new optical probe to study norepinephrine synapses in vivo. In this work we were able to observe the effects of amphetamine on norepinephrine release in vivo and observed some evidence of potential silent synapses. I also describe a new method of cranial window surgery I developed for optical imaging. This technique called PHASOR, is faster, and has a higher success rate, than traditional surgical methods. The improvements demonstrated in this new surgical technique may enable more widespread use of optical imaging methods. In the second part of my thesis, I used optical techniques to study the dorsal striatum in vivo in awake behaving mice. The direct and indirect paths of the dorsal striatum play an important role in motor behavior and motor learning. Dysfunction in these paths has been implicated in motor diseases as well as in mood disorders. In this thesis, I provide a review of the anatomy and physiology of the neurons that comprise the dorsal striatum, and the circuits that they form. The next chapters describe my work using optical techniques to record from these neurons in vivo. In my first set of experiments, I recorded from the direct and indirect paths during a behavioral task of anxiety and observed differential firing depending on the anxiety state of the mouse. Finally, in a preliminary set of experiments, I record from the direct and indirect paths during tasks of motor learning. I found that both paths show changes in firing during motor learning and that these changes differ between the dorsolateral and dorsomedial striatum.