545298 research outputs found
Sort by
Anxiety Effects on Decsion Making with College Students
Get PDFWith rising levels of anxiety among college students, understanding its effects on decision-making is crucial, such as whether to enter the workforce or attend graduate school. In two studies with undergraduate business students, we examined the correlation between anxiety and decision making (Study 1; N = 204) and experimentally induced anxiety to test for causality (Study 2; N = 249). We assessed decision-making using a hypothetical choice scenario between an entry-level job offer versus an MBA program, as well as standard measures of decision-making traits: time preference, risk aversion, and loss aversion. We hypothesized that higher anxiety levels would lead to greater preference for immediate rewards, greater risk and loss aversion, and thus greater preference for the job over the MBA. Counter to our hypotheses, neither study found a significant relationship between either measured or induced anxiety and participants’ decision between the job and the MBA. In addition, our anxiety manipulation did not work as planned in Study 2, with no differences in self-reported anxiety between conditions. We also did not find relationships between anxiety and either risk or loss aversion, but did find that anxiety correlated with more impatient time preferences in Study 1. Participants in the experimental condition had marginally more impatient time preferences and greater loss aversion than those in the neutral condition in Study 2. Future research should explore alternative methods of anxiety induction and other student populations to better understand how anxiety changes student decision making
Hippocampal Structure and Function are Modulated by a Dynamic Endocrine Environment
Get PDFHistological evidence suggests that the estrous cycle exerts a powerful influence on CA1 neurons in mammalian hippocampus. Decades have passed since this landmark observation, yet how the estrous cycle shapes dendritic spine dynamics and hippocampal spatial coding in vivo remains a mystery. To address this, we developed a custom hippocampal microperiscope and used two-photon calcium imaging to track CA1 pyramidal neurons in female mice across multiple cycles. Estrous cycle stage was determined using a custom deep neural network (EstrousNet) to automate classifications of vaginal cytology. Estrous stage had a potent effect on spine dynamics, with spine density peaking during proestrus when estradiol levels are highest. These morphological changes coincided with greater somatodendritic coupling and increased infiltration of back-propagating action potentials into the apical dendrite. Finally, tracking CA1 response properties during navigation revealed greater place field stability during proestrus, evident both at the single-cell and population level. Future studies will leverage conditional knockdowns of estradiol receptors in hippocampus and tracking of behavioral syllables during navigation to determine the molecular and behavioral dimensions of estrous modulation. Additionally, neuronal morphology in CA1 will be tracked across other hormonal events such as pregnancy and postpartum. Taken together, these findings demonstrate that the estrous cycle drives large-scale structural and functional plasticity in hippocampal neurons essential for learning and memory
Citizen Engagement and Policy Change
No full textThis dissertation considers the various political forces that shape citizen participation in public policymaking in general, and in climate policy debates in particular. The first chapter examines constituent contacts to Congressional offices. It asks, how might inequalities in who contacts Congress, and on which issues, bias the information that legislators and their staffers receive about public opinion in their districts? This chapter leverages a novel dataset of administrative contact data, paired with district-wide public opinion surveys, to uncover demographic, organizational, and reactive biases in citizen engagement with the U.S. Congress. The following chapters examine these forces within U.S. climate and energy politics, an area where federal policy outcomes have historically been misaligned with public opinion. The second chapter analyzes the development and enactment of the Inflation Reduction Act of 2022, drawing on legislative, archival, and insider accounts. It interrogates the role of the climate movement, entrenched interests, and the status quo in influencing the policy’s shape and fate. Finally, drawing on theories of policy feedback and political mobilization, chapter three uses a novel survey experiment to explore citizen responses to the threat of policy repeal. 
Disciplinary Negligence: At-Promise Latino Youth and The Disciplining of Support in a “Non-punitive” Era
No full textDisciplinary Negligence is a three-year school ethnography exploring the experiences of at-promise Latinx students as they navigate the dual processes of school discipline and support networks. Importantly, this school district revised its discipline policy in 2021 to minimize suspensions and expulsions providing a unique opportunity to observe organizational change and student outcomes. From 2021-2023, I completed participant observations in the dean’s office, disciplinary intervention room, classrooms, hallways, and off-campus lunch sites frequented by students. I supplement my ethnography with 60 semi-structured interviews of students, staff, and administrators informed by my observations to support this study’s findings. Building on the work of scholars such as Jonathan Simon and Victor Rios who variously have shown that public schools in racially segregated communities are places of punitive social control, I studied a diverse suburban high school on the Central Coast of California that implemented a non-punitive disciplinary policy reform and ultimately uncovered a distinct process that synthesizes punishment with neglect. This process, which I call “disciplinary negligence,” explains how practices of restorative social control within the school obscure and promulgate neglect of at-promise Latinx students. My findings extend scholarship beyond theories of punitive social control to consider how neglect functions as an essential mechanism for maintaining what is perceived to be a productive school.Across three empirical chapters, I develop my theoretical framework of the “disciplinary negligence” and examine its impacts on at-promise Latinx youth. In Chapter 2, I draw on observations and interviews with school staff to argue that institutional reluctance to shift away from punitive social control is rooted in a cultural frame that prioritizes safety and intensifies focus on perceived school dangers. Chapter 3 illuminates the school’s attempts to establish a non-punitive culture, and how they were limited by the prioritization of managing safety issues. Although, suspensions and police referrals reduced, the school’s hyperfocus on danger undermined trust between institutional agents and at-promise Latinx students—many avoided seeking support due to fears of punishment, which undermined restorative justice efforts and reinforced their neglect. In chapter 4, I expose the barriers at-promise students continue to have to overcome on their path to graduation. Many of these barriers were influenced by the lack of engagement teachers and staff had with these students. Also, staff attempted to limit their access to institutional agents and high-achieving peers as part of the disciplinary process. I conclude with a summary of findings, policy recommendations, and broader sociological implications. This study contributes to research in Latinx sociology, inequality, and the sociology of education by documenting how even with well-meaning changes to school discipline, these processes still reinforce neglect and emotional disengagement from education. My findings deepen understanding of durable inequalities amid equity-driven policy changes especially as it pertains to marginalized youth
A Thurston Compactification of Stability Manifolds for some Local Calabi-Yau Varieties
Get PDFThis dissertation presents a number of results on the partial compactification of the Bridgeland stability manifolds associated to a few well-behaved complex varieties by considering their embedding in a certain real projective space. The first main result demonstrates that encoding stability conditions by the masses of a large set of objects always yields an embedding for a significant number of geometric examples — in almost all cases, however, this yields an embedding into an infinite-dimensional space. The second main result shows that a large class of geometric examples proposed by fail to admit a Thurston compactification. The last result is a description of the boundary of the compactification for a certain non-compact Calabi-Yau surface; this can be seen as a geometric realization of the original Thurston compactification for the A2-quiver introduced by Bapat, Deopurkar, and Licata. This leads to a similar approach in one dimension higher with a much more complicated boundary structure
Three Studies in Experimental Economics
Get PDFThe three chapters of this dissertation present experiments that explore the demand for verification in the context of misinformation, how experience affects risk attitudes, and citizens’ decision to become candidates in elections. The variety of topics covered in the present dissertation attests to the flexibility and applicability of experimental methods in economics and the insights we can gain from them.In the first chapter, participants had to classify if a headline was accurate or contained false information and report their willingness to pay (WTP) to verify it. This experiment controls for the payoffs participants receive for a correct classification, the accuracy of the verification process, and the prior probability of observing misinformation. All these variables are relevant in the decision-making process to verify headlines. However, these variables are impossible to control and often challenging to measure in the field. This experiment, conducted in Mexico, used real headlines and monetary incentives to measure the demand for verification, accuracy rates, and confidence. Participants had a larger WTP, accuracy rate, and confidence when they classified political headlines. Prior headline classification as true strongly affected the WTP. Classifying a headline initially as true also increased the WTP and accuracy among political headlines. Finally, a negative interaction effect was found between headlines favoring the government and the participants’ criticism of the government.In the second chapter, the question was: Do people know their own risk preferences, or do risk choices change with experience and observation? A straightforward test in the laboratory is provided. People make an initial decision concerning a lottery choice and then experience 24 unpaid practice periods in which they roll the dice, record the outcome, and record the would-be payoff. They then make a final decision for the lottery choice; one of the first and last periods is randomly chosen for payment. The primary hypothesis was that people will become less risk-averse by having made and experienced the practice rolls. People are significantly more likely to become less risk-averse than more risk-averse over time. This move towards assuming increased risk goes in the opposite direction from what is at least arguably predicted by loss aversion and reference dependence. Also, it was found that women’s preferences changed much less than men’s. We feel that our hands-on approach ensures a degree of engagement that helps to accelerate the learning process. It is argued that measures obtained after people have had experience with a mechanism are more meaningful, and this principle might extend more generally to other elicitation tasks.In chapter 3, the findings were reported from a study that explores candidate participation in a context where citizens can become candidates under both plurality and run-off voting systems. The study also considers the influence of entry costs and different platforms of potential candidates. While the findings align with the expected outcomes of the citizen-candidate model, there is a notable over-participation by candidates from less favorable electoral positions. These entry patterns adjusted well to the QRE. This research adds to the existing body of knowledge about what motivates candidates to enter races under different voting systems and analyzes the behavior of candidates in extreme positions
Programming the Cosmic Time Machine: New Ideas and Applications of Machine Learning For Stage-IV Spectroscopy and Beyond
Get PDFThe nature of Dark Energy is one of the most significant and compelling unsolved problems in physics today. Stage-IV spectroscopic surveys like the Dark Energy Spectroscopic Instrument (DESI) aim to try probe this unknown sector of physics using observables like Baryon Acoustic Oscillations (BAO). DESI recently released its Year Three results, with some of the tightest constraints on the composition of the universe recorded to date.DESI uses a state of the art software and data processing pipeline that makes significant use of a variety of machine learning algorithms. I present in this work a selection of four different machine learning and algebraic techniques developed for use as part of the DESI survey. The first algorithm applies deep learning to the problem of cosmic ray identification and rejection. Cosmic rays are a persistent source of error when extracting spectroscopy from raw CCD images and thus necessitate an accurate and fast algorithm for detection and masking.The second algorithm is QuasarNET, a deep convolutional neural network designed to automatically classify quasars in raw DESI spectra. QuasarNET also estimates a quasar redshift for each quasar identified. In my work we retrain QuasarNET using an algorithm called Active Learning, which automatically determines which unlabeled spectra would be beneficialto label. We then use those newly labeled spectra as a training dataset for QuasarNET, in the process discovering and later fixing a systemic problem with QuasarNET redshift estimates. This new weights file was used for DESI Year Three analysis and will be used in Year Five and beyond.The last two algorithms are centered on linear algebra and matrix decomposition. I present new unified coaddition scheme called “Bayesian Coaddition” that unifies three different coadds into a single Bayesian likelihood with a single hyper parameter prior. This work includes a coadd that reconstructs the true image without any telescope transmission effects, as well as a coadd with a diagonal covariance that is the statistically optimal way to weight exposures with different seeing when coadding. Finally I present an algorithm for non-negative matrix factorization that generates non-negative decomposed matrices of templates and coefficients that does not require the input data to be non-negative
Semiclassical Quantum Gravity and Holography
Get PDFThis thesis reports on several aspects of semiclassical quantum gravity within the framework of holography duality. Among these are the relation between boundary causality and locality in AdS/CFT, correlation functions in 3d gravity and 2d CFT, and spin-refined observables computed by the gravitational path integral. Chapter 1 is introductory and describes some general features of semiclassical quantum gravity, holographic duality, and universal aspects of AdS3/CFT2. Following this introduction, Chapter 2 develops the machinery of 2d CFT more thoroughly. Chapter 3 studies an apparent paradox in AdS/CFT whereby the boundary subregion encoding a bulk point can appear to violate causality by moving superluminally. The paradox is resolved by the non-local nature of the encoding, which gives the boundary subregion its finite extent. The encoding region cannot jump outside of its domain of dependence, and therefore remains in causal contact with itself as it moves.Chapters 4 and 5 are concerned with holographic correlators in AdS3/CFT2. We consider fields propagating in the geometries of heavy objects like conical defects and BTZ black holes, which have dual descriptions as excited or thermal states of the CFT. These chapters describe several calculations of correlators of such fields using both gravity and CFT techniques. The match between the bulk and boundary results allows us to extract universal CFT data that determines the amplitudes to create excitations of the light field, as well as the expectation values of an infinite family of two-particle states exchanged between the heavy and light fields. In the CFT, these are HHL OPE coefficients for a family of double-trace operators whose presence is required by crossing symmetry.Chapter 6 investigates spin-refined partition functions in AdS/CFT using the Euclidean gravitational path integral. These objects have the form Tr (e −βHX), where X is a discrete isometry. We construct phase diagrams for various choices of X, devoting particular attention to the case where X is a reflection. A CRT -twisted black hole universally dominates the canonical ensemble for this quantity at high temperatures, while complex rotating black holes dominate the microcanonical ensemble at high energies
Symmetry and Universality in Quantum Many-Body Systems
Get PDFThis dissertation reports on recent novel results regarding symmetry and quantum critical phenomena in condensed matter and quantum many-body systems. Such systems, composed of many strongly-interacting degrees of freedom, are classified by their long-distance behavior captured by fixed-points of the Renormalization Group (RG) in a Quantum Field Theory (QFT) framework. Additionally, this universal long-distance physics is heavily constrained by generic properties of the system such as symmetries and their potential ’t Hooft anomalies. In particular, we focus on two different kinds of such physics in this thesis.This text can be largely divided into two parts. The first part, which includes chapters 2 through 4, discusses statistical and quantum lattice models with “subsystem symmetries”. Such symmetries come in two flavors: type-I, where the symmetry charge is supported on a regular submanifold and type-II, where the support of the symmetry charge is fractal instead. We explicitly construct quantum lattice models with type-II (or, fractal) subsystem symmetries composed of qudit or rotor degrees of freedom. These models may support gapped phases that spontaneously break fractal subsystem symmetries. A multicritical point endowed with such symmetry will be remarked upon in Chapter 3 and correlation functions in the ordered phases and at the multicritical point will be also discussed. An experimental proposal for realizing similar models in experimental platforms of Rydberg atoms will also be considered.The second part of this dissertation – contained in chapters 5 through 8 – will instead focus on classifying universal physics in open quantum systems, particularly those coupled to ancilla degrees of freedom without dynamics or a Markovian bath, for finite-time. These effects are realized as decoherence (or weak-measurement) acting on the original system, transforming the quantum wavefunction into a mixed state. The long-distance scaling of entanglement measures and other information-theoretic quantities may see marked change, signifying such mixed-states are examples of new unconventional quantum matter. An array of quantum many-body systems subjected to these effects, including quantum critical points and Chern insulators, will be discussed
A survey on DNN-based PDE solver methods
Get PDFPartial differential equations (PDEs) are central to modeling physical, biological, and engineered systems, yet their analytical solutions are rare, and numerical methods often face scalability and efficiency challenges in high dimensions or complex domains. Recent advances in deep learning have led to the development of neural network-based PDE solvers, including physics-informed neural networks (PINNs), the Deep Ritz and Deep Galerkin methods, and neural operator frameworks such as Fourier Neural Operators (FNOs) and DeepONets. These approaches offer mesh-free, differentiable alternatives to classical solvers and demonstrate promise in handling high-dimensional, multi-physics, and data-sparse problems.This survey presents a comprehensive review of deep neural network (DNN)-based PDE solvers, beginning with foundational variational and residual-based formulations, and progressing to operator learning approaches that generalize across input function spaces. We compare algorithmic frameworks, training paradigms, and performance characteristics, and highlight applications across physics, engineering, and biology. Key implementation challenges—such as enforcing boundary conditions, ensuring convergence, and scaling to high-dimensional settings—are discussed alongside recent strategies to mitigate them. Finally, we explore emerging directions, including hybrid solvers, uncertainty quantification, and scientific foundation models. This survey aims to provide researchers and practitioners with a structured overview of the field and guidance for future development of scalable, data-integrated, and physically consistent PDE solvers