36214 research outputs found
Sort by
Everywhere the Light Touches / Alice Through the Looking Glass: Exposing the History of a Well-Behaved Woman
No full textThesis (Master's)--University of Washington, 2025When history spotlights the man at the center, who is left in the shadows? Everywhere the Light Touches, a novel, reilluminates the life of pioneering businesswoman Alice Kay Whitney Hutchison, longtime secretary to George Eastman, founder of Kodak. Written in an experimental, epistolary form that blends omniscient and first-person narration, it imagines Alice’s response to Eastman’s suicide as she questions their life’s work in light of the compromises they made to protect their private selves. A lens of grief and memory invites readers to reconsider Alice’s role in the intersecting histories of photography, capitalism, and feminism at the turn of the twentieth century. The accompanying essay, “Alice Through the Looking Glass: Exposing the History of a Well-Behaved Woman,” draws on the work of Laurel Thatcher Ulrich to examine the politics of archival silence and the ethics of recovering “well-behaved” women from historical obscurity. Taken together, the two pieces offer a feminist reimagining of erased labor and overlooked lives, while probing the unstable boundaries between fact and fiction, past and present
Physician Performance in the Merit-Based Incentive Payment System: Implications for Health Disparities
No full textThesis (Master's)--University of Washington, 2025The Merit-based Incentive Payment System (MIPS) is ostensibly designed to promote better care across participants. However, MIPS risks exacerbating disparities among participants with fewer resources to invest in performance metrics like care delivery improvements or quality reporting. Unfortunately, little is known about how group practice characteristics impacted MIPS scores and associated reimbursement in later years of implementation. Two practice-level characteristics (location in urban areas and larger patient populations) and two community-level characteristics (some college education and health care spending) were positively associated with MIPS scores. In contrast, patient population case mix and the proportion of Medicare/Medicaid dual-eligible patients were negatively associated with MIPS scores at the practice-level. These findings underscore the potential risk that MIPS may exacerbate health disparities by penalizing practices caring for lower-income populations adversely affected by social drivers of health
“Definite Renewal”: Emotional and attention impacts of rest spaces in museums
Get PDFThesis (Master's)--University of Washington, 2025Dedicated areas for visitors to rest, recharge, and reflect on their experiences are rare in museums, but such features are becoming more common and have been included by several notable institutions. Currently, there is little research exploring how these spaces impact museum visitors. The purpose of this survey study was to explore what drew adult museum visitors to use dedicated rest spaces, and how time spent in them impacted visitors’ emotional states and attentiveness. Sixty-eight adults were interviewed at two museums after spending three minutes or more in a dedicated rest space. Many visitors reported feeling calm during their time in the rest spaces, and cited specific design and curatorial choices that contributed to this impact. About a quarter of adults reported that they felt their attention had improved after their rest time; another quarter cited restorative impacts unrelated to attention; and the remaining half were uncertain or reported no change. These findings indicate that dedicated rest spaces have the potential to foster feelings of calm and increase attentiveness, which could help facilitate learning in museums
Designing for Communication, Collaboration, and Coordination in Accessibility and Health
Get PDFThesis (Ph.D.)--University of Washington, 2025Communication, collaboration, and coordination are fundamental to human interaction. As computing increasingly mediates these interactions, it is essential to address the sociotechnical challenges that arise in complex settings, such as accessibility (e.g., interactions among people with and without disabilities) and health (e.g., patients with chronic conditions collaborating with providers for care). This dissertation focuses on the design, implementation, and examination of two systems: (1) Jod, an accessible videoconferencing platform tailored to facilitate communication in mixed hearing groups, and (2) SCOPE, a technology-enhanced platform for collaborative care management for patients with cancer and depression. Grounded in the design and domain theories that informed the development of these feature-rich platforms, this work critically examines these systems through a sociotechnical lens. The complexities of settings like accessibility and health reveal a range of needs for communication, collaboration, and coordination, motivating sociotechnical systems that can adapt according to evolving social dynamics
An Assessment of an Agri-environmental Policy on Landscape Connectivity for Wildlife
No full textThesis (Master's)--University of Washington, 2025Agricultural land conversion has resulted in the loss of 95-98% of tallgrass prairie habitat in the US. The rapid expansion and intensification of agricultural practices has played a pivotal role in meeting the ever-growing global demand for food. However, this intensification has come at an undeniable cost to the environment, especially in the form of habitat loss. In a high intensity agricultural system such as the US corn belt, agri-environmental policies can have a significant impact on remnant ecosystems. One common policy in this regard is the implementation of riparian buffer strips, designed to reduce pollution and soil runoff from agricultural fields into adjacent waterways. However, these strips also have the potential to provide critical habitat for wildlife, fostering landscape connectivity and aiding in the conservation of biodiversity. While the federal government provides incentives to encourage the creation and maintenance of buffer strips, the only state in the corn belt to require them is Minnesota. I built a machine learning model to classify high-resolution aerial imagery of this region over a time series from 2008, before policy implementation, to 2021, after implementation. The model separated agricultural and developed lands from permanent vegetation with the assumption that permanent vegetation will support the dispersal of wildlife more effectively than corn and soy monocrops. These classified maps were quantified using landscape connectivity metrics and compared between MN and other corn belt states, which collectively served as a synthetic control. The synthetic control model estimated high, positive impacts of buffer strip policy on landscape connectivity metrics
Influenza A Virus Pathogenesis in a Non-Human Primate Model of Pregnancy
No full textThesis (Ph.D.)--University of Washington, 2025Influenza A viruses (IAVs) have caused pandemics throughout the last 100+ years and remain a danger to global health due to their evolutionary mechanisms and persistence in worldwide avian populations. Pregnant women infected with pandemic IAVs have greater risk of respiratory disease severity and pregnancy complications. We hypothesized that aberrations in the pregnant primate's innate immune response to pandemic IAV predisposes them to severe IAV lung disease and atypical placental function. We infected pregnant and non-pregnant non-human primates (NHPs) with 2009 H1N1 and euthanized them at 5 days post-infection. Pulmonary physiology tests showed that severe IAV respiratory disease emerged at a similar frequency in both pregnant and non-pregnant macaque females. We observed that upregulation of antiviral gene expression in the lungs was similar in both infected pregnant and non-pregnant females relative to uninfected females. However, frequencies of IAV-infected alveolar macrophages are significantly higher in pregnant macaques relative to non-pregnant females. A multivariate linear regression model suggests that pulmonary TLR9 expression combined with baseline oxygenation values predict endpoint disease severity, regardless of pregnancy status. We did not find evidence of viral infection in most placentas or any fetuses. We encountered no significant differences in frequency of injury between placentas from infected and uninfected NHPs. In spite of this, we found evidence of elevated antiviral gene expression and pro-inflammatory cytokine circulation in placentas from infected versus uninfected NHPs. We calculated that maternal lung viral load correlated significantly and positively with antiviral gene expression, suggesting that the extent of maternal infection is triggering placental innate immunity. We believe that our translational NHP model reveals interesting insights into the maternal innate immune response to IAV infection and crosstalk with placental immune activation. Future studies should use longer experimental timelines to quantify enhanced mortality and pregnancy complication risks from maternal IAV infection
Impacts of interannual isoprene variations on methane lifetimes and trends
No full textThesis (Master's)--University of Washington, 2025Methane is a potent greenhouse gas that has an atmospheric lifetime of 9-12 years due to its reaction with the hydroxyl radical (OH). Recent trends in methane indicate anomalously high methane growth in 2020, which has been attributed to a combination of increased wetland emissions and a decrease in OH from decreased anthropogenic nitrogen oxide (NOx) emis-sions during the COVID-19 lockdowns. However, NOx is not the only atmospheric species that affects global OH concentrations—isoprene, the most significant non-methane hydrocarbon by total emissions, is oxidized by OH, which can deplete OH during periods of high isoprene emissions. Using satellite isoprene retrievals from the Cross-track infrared sounder (CrIS) instrument, we find interannual variability in isoprene columns and anomalously high isoprene columns during 2020, coincident with anomalously high methane growth. Although isoprene is concentrated in source regions such as the Amazon, its atmospheric oxidation produces carbon monoxide, which can be transported over longer distances and thus impact the global atmospheric oxidative capacity. Elevated isoprene concentrations may have contributed 13% (bounds: 10% - 28%) of the total methane growth in 2020 if we assume no change in NOx emissions between 2019 and 2020. Given that COVID-19 lockdowns also decreased anthropogenic NOx emissions during this same period, this estimate of isoprene emissions on the methane lifetime during 2020 is an upper-limit, and the resulting change in the oxidative capacity may be sensitive to whether isoprene emissions or NO emissions drove this isoprene anomaly. This study illustrates how the biosphere may impact the atmospheric oxidative capacity and, in turn, the methane lifetime through interannual changes in biogenic emissions that are forced by meteorological and climate variability
Understanding the Effects of Human Recreation on Wildlife from Fieldwork to Management Decisions
No full textThesis (Master's)--University of Washington, 2025Interest in outdoor recreation is increasing across natural landscapes, benefiting people and communities, but posing challenges for wildlife and land managers. This thesis examines recreation and wildlife management through field-based research on wildlife responses to recreation and an exploration of how scientific research informs management decisions. In Chapter 1, we examined the impacts of recreation on wildlife using camera trap data from 113 stations across eight Washington State Parks in the wildland-urban interface (WUI) — critical ecological zones where recreation-wildlife dynamics remain understudied. Using dynamic occupancy models and temporal activity analyses, we found that recreation influences species detection probabilities and activity patterns. Raccoons, bobcats, and cougars were more likely to be detected at stations with high recreation, while black bears, elk, black-tailed deer, and coyotes had higher detection probabilities at stations with low recreation. Temporal shifts were evident among coyotes and deer, which had significant differences in their activity patterns, showing increased diurnal activity in off-trail areas. To understand how field-based studies inform recreation and wildlife management, Chapter 2 examines the role of science in decision-making. We conducted semi-structured interviews with 14 land managers from local, state, and federal agencies in Washington, using thematic analysis grounded in the Theory of Planned Behavior. While managers expressed positive attitudes toward science-based decision-making, structural constraints, resource limitations, and communication gaps were barriers to research integration. In contrast, partnerships and accessible science were key facilitators, with tribal and local community pressures shaping management priorities. As outdoor recreation expands, maintaining spatial and temporal refuges for wildlife and strengthening the integration of science into management decisions will be essential for balancing wildlife conservation with sustainable recreation access
Developing techniques for Simulation of SU(3) Quantum Field Theories on State-of-the-Art Quantum Devices
No full textThesis (Ph.D.)--University of Washington, 2025Quantum computing has long been an experimental technology with the potential to enablesimulation at scale of phenomena which on classical devices would be too expensive to simulate
at any but the smallest scales. Over the last several years, however, it has entered the
NISQ era, where the number of qubits are sufficient for quantum advantage but substantial
noise on hardware stands in the way of this achievement. This thesis details improvements
to techniques of quantum simulation of the out-of-equilbrium real-time dynamics of lattice
quantum chromodynamics (LQCD) and of dense 3-flavor neutrino systems on digital quantum
devices that I have contributed to as a Ph.D. student.
LQCD has been numerically computed using classical Monte-Carlo methods for decades
with applications to nuclear and particle physics. However, classical Monte-Carlo methods
experience the numerical sign problem, which limits their ability to solve problems such as
the simulation of out-of-equilibrium real-time evolution. These problems are an ideal place to
look for quantum advantage. The first project in this part is a comparison of gradient descent
and the Bayesian process as choices for the classical optimizer within a variational quantum
eigensolver that initializes the ground state of an SU(3) plaquette-chain. The thesis then
pivots to a 1+1D lattice of quarks interacting with an SU(3) gauge-field. A VQE-based statepreparation
for the vacua and a Trotterized time-evolution circuit is designed and applied to the problems of simulating beta and neutrinoless double beta decay. The latter has been
implemented on Quantinuum's H1-1 trapped ion device. Finally, these circuits are adapted
to a version useable on quantum devices with nearest-neighbor connectivity with minimal
overhead, with an eye towards utilizing the higher qubit count of such devices for hadron
dynamics and scattering.
Physical dense neutrino systems are both highly out-of-equilibrium and are characterized
by a highly entangling interaction, neutral current exchange between neutrinos. This makes
the dynamics of dense neutrino systems, particularly the relatively not-well-studied 3-flavor
variety, an ideal problem to attempt to solve with quantum computers. In this part, Trotterization
circuit-elements for the neutrino-neutrino interactions that happen in ultradense
3-flavor neutrino systems are designed and implemented on Quantinuum's H1-1 trapped ion
device and IBM's ibm torino superconducting device. A circuit for Trotterizing this interaction
on qutrits, which is substantially lower in depth than its qubit counterparts, is also
created. Lastly, by the Gottesman-Knill theorem problems need to exhibit both high entanglement
and high deviation from stabilizer states ("magic") in order to exhibit quantum
advantage. Thus, the second-to-last chapter of the thesis detail results with implications
for the Standard Model in general that the 3 flavor ultradense neutrino systems with the
highest, most-persistent magic are those that start with neutrinos in all 3 flavors
Disparities in Autism Spectrum Disorder Diagnosis: Examining Individual and Socioeconomic Predictors of Diagnostic Age and Provider Type
No full textThesis (Ph.D.)--University of Washington, 2025Autism spectrum disorder (ASD) can be reliably diagnosed as early as 18-months-old, however, the average age of diagnosis is 48 months or later. It is critical that individuals suspected of having ASD are evaluated as early as possible, as early intervention has been shown to have the most robust long-term outcomes. Individual and systemic barriers prevent underserved populations from receiving a timely diagnosis of ASD. While these disparities have been outlined in the literature, to date, a large-scale study has not examined the impact of these individual predictors, along with access to resources, has on the timeliness of diagnosis. As such, the present uses linear modeling with data from the SPARK study to examine how individual and socioeconomic predictors impact the timing of diagnosis, particularly for participants from traditionally underserved groups. Additionally, the present study will examine the role of school psychologists in the identification of ASD based on individual and socioeconomic factors. To achieve this, logistic regression was used to model the probability of school-based ASD diagnosis. Results indicate that cognitive impairment, symptom severity, diagnosis year, and BIPOC status were predictive of diagnosis age. Additionally, interactions with cognitive impairment and symptom severity were significant. For school-based diagnoses, BIPOC individuals were more likely to be diagnosed, while females were less likely. Age band and diagnosis year also interacted with the likelihood of receiving a school-based diagnosis. These findings highlight the importance of addressing both individual and socioeconomic factors that contribute to diagnostic age for ASD, particularly for underserved populations. The results suggest that greater attention is needed to ensure equitable access to timely evaluations and interventions, especially for BIPOC individuals and females, who are at risk for later diagnoses despite early indicators of ASD. Additionally, the role of school-based identification is crucial, as individuals from undeserved communities are more likely to receive diagnoses in the school setting