22689 research outputs found
Sort by
REFINEMENT OF MURINE AND NON-ANIMAL MODELS TO IMPROVE TRANSLATIONAL NEUROTOXICITY RESEARCH
Get PDFNeurodevelopmental disorder research faces significant translatability crisis. Indeed, the NIH reports that approximately 90% of pre-clinical neuroscience research fails to progress into clinical trials, and of those that enter, face a 97% attrition rate in Phase III. This translational gap remains attributed to multiple factors, namely: the complex genetic and environmental etiologies, species-specific neurocognitive difference, the unintentional biasing of animal model through poor husbandry and handling, and a dearth of valid human-relevant models. This thesis aims to address these challenges by refining both animal and hiPSC-derived organoid models to enhance translational outcomes.
The first contribution investigates how early-life events, specifically anesthesia, impact neurocognitive and motor skills in rodents. C57BL/6 murine pups were exposed to isoflurane during the critical developmental window of synaptogenesis. Cardiorespiratory and arterial blood gas parameters were measured to assess the physiological impact. Subsequently, behavioral tests were conducted on adult mice previously anesthetized as pups. Results revealed that exposure to isoflurane for over 3 hours engendered profound acid-base and respiratory disturbances. Additionally, early anesthesia induced sex-specific neurocognitive and motor changes in adulthood, specifically females displayed short-term memory deficits, and males exhibited increased anxiety and reduced gross motor activity.
The second and third contributions investigate how hiPSC-derived organoid models may be improved to better recapitulate the human brain microenvironment. First, we established an age correlation for organoids, as compared to mice, and humans, through the identification of a well-defined developmental biomarker: the GluN2B-GluN2A. We found it occurred at 6 weeks post-differentiation in organoids, between PND7-11 in mice, and six months in humans. The final contribution aimed to address a major criticism of hiPSC-organoids, namely their lack of microglia. We established a novel protocol for integrating hiPSC-derived microglia into organoids. As compared with standard organoids, the microglia-integrated model demonstrated enhanced neuronal and glial maturation, increased synaptogenesis, and immune functionality through responses to antigen challenge and exhibition of activated morphology phenotypes.
In summary, this thesis addresses the translatability crisis by refining both animal and hiPSC-organoid models. Collectively, the enhancement of pre-clinical models through continued refinements will strengthen the validity and translational value, which will ultimately enable more efficacious scientific outcomes
THE MALE EPIDEMIC: EXAMINING WHY FIRST-YEAR STUDENTS ARE DROPPING OUT OF COLLEGE
Get PDFCollege dropout rates have increased worldwide over the past decade as first-year male students are dropping out at higher levels than those of their female classmates. This disparity stems from many sources, including gendered parenting, neurobiological differences between boys and girls, prevalence of neurodevelopmental disorders in boys, peer groups, and overconfidence. While financial challenges remain the primary reason why students drop out of college, this study delves into other causes, such as executive functioning deficits, limited emotional regulation, and social isolation. This study uses Bronfenbrenner’s ecological systems theory as a lens to examine factors that influence students at college and Albert Bandura’s theory of self-efficacy as a theoretical framework. Meta-analysis is incorporated to examine how self-efficacy impacts student outcomes at college on a worldwide basis
Bondservants in the Imperial “Homeland”: Gender, Service, and the Making of the Early Qing, 1616-1722
No full textThis dissertation examines how the regulation of bondservant women in Liaodong functioned as both a mechanism of imperial governance and a site of local resistance in the early Qing dynasty. I argue that gendered governance was not peripheral but fundamental to Qing statecraft—a sophisticated system through which the empire produced political subjecthood, established hierarchies, and extended administrative reach into frontier regions. Rather than treating subject status as pre-existing cultural identity or inherited social rank, this study reveals it as a political construct systematically produced through institutional practices, documentary regimes, and corporeal discipline.
Rooted in Jurchen traditions of mutual dependency, this mode of governance rendered women especially vulnerable to imperial intervention. Under Nurhaci, women’s allocation functioned as an instrument of personalistic rule, reinforcing hierarchical bonds of loyalty. Hong Taiji extended this logic by explicitly targeting marriage as a site of regulation, while the Shunzhi emperor introduced gendered population registration policies. The selection of xiunü policy, further refined by Kangxi, brought this logic to its most intimate frontier, enabling the state to govern women’s bodies—determining the timing of female adulthood and parameters of legitimate marital alliances.
My analysis of bondservant women reveals how administrative documentation—not ethnicity or class—determined obligations to the state. Many girls remained in “bureaucratic suspension,” revealing tensions between imperial ambitions and household needs. After the 1680s, selection policy evolved into a distinctive mechanism of frontier control in Mukden, while marriage regulation emerged as a sophisticated tool for social engineering. Illicit marriages—particularly those crossing banner boundaries—became sites where families, officials, and central authorities negotiated status and identity. A key transformation occurred in the late 1670s-1680s: the reorganization of bondservant administration followed by expansion of imperial estates, shifting enforcement from fragmented responses to systematic processes.
Through archival analysis, I demonstrate that the northeast functioned not as a passive periphery but as a laboratory for governance experiments. This study advances a new framework for understanding early Qing statecraft—one operating simultaneously at the levels of body, household, and frontier space, revealing governance not as unidirectional imposition but as continuous negotiation between imperial ambitions and local realities
Essays on Inflation and Exchange Rates
No full textThis dissertation presents three essays on inflation and exchange rates.
The first chapter re-evaluates whether the vacancy-to-unemployment ratio (V/U) is a more accurate measure of labor market tightness than the unemployment rate (U). To address concerns that V/U may over-fit the COVID-19 episode, I use pre-pandemic time-series data and exploit better identification from Metropolitan Statistical Area (MSA)-level panel data. I construct state-space models and apply the Kalman filter to MSA-level panel data, jointly estimating a non-linear Phillips curve and time-varying natural rates of U and V/U. While time-series data cannot distinguish between the two measures on pre-pandemic data, panel data indicates that V/U performs better even before the pandemic. I also show in a theoretical model that V/U is the right variable in the Phillips curve, and V/U is more informative than U when the Beveridge curve shifts.
The second chapter studies why the national Beveridge curve shifted outward significantly during the COVID-19 period (2020M4–2022M4) compared to the pre-pandemic era (2009M7–2020M3). We introduce a novel metric to quantify curve shifts based on the V = U intersection, and leverage cross-sectional variation across industries and states. We document substantial heterogeneity in sectoral shifts, with larger shifts in low-earning sec- tors and those with limited telework capacity. These findings can be related to low-earning sectors having a stronger reduction in search from higher unemployment insurance (UI), and telework capacity serving as a match-preserving technology. At the state level, shift magnitude depends on pre-pandemic industry composition and COVID-19 infection rates. The results suggest that expanded UI, limited telework capacity, and COVID-19 severity jointly drove the aggregate Beveridge curve shift during the pandemic.
The third chapter examines the effect of geopolitical risk (GPR) and trade policy uncertainty (TPU) shocks on exchange rates using a panel local projection framework. We analyze monthly bilateral exchange rates of advanced and emerging economy currencies against the U.S. dollar from 1985M1 to 2025M2. We find that global GPR shocks depreciate advanced economy currencies, while country-specific shocks depreciate emerging market currencies. TPU shocks lead to depreciation against the U.S. dollar in both groups
CULTIVATING CONNECTIONS: THE ROLE OF STUDENT-TEACHER RELATIONSHIPS IN FOSTERING CREATIVITY IN SCIENCE EDUCATION
Get PDFThis mixed-methods study explores the vital role of student-teacher relationships in promoting creativity in science education, addressing ongoing challenges in American science achievement that have lasted since the Space Race era. The research examines how meaningful educator-student connections act as catalysts for developing creative thinking skills essential for 21st-century scientific literacy. Grounded in Bronfenbrenner's ecological systems theory, this study used a networked approach to explore factors at various system levels that influence creativity in science education. The research focused on three main questions: teachers' current awareness of creativity in science education, instructional strategies that promote scientific creativity, and how cultural responsiveness affects teachers' ability to foster creativity. Data collection included semi-structured interviews with science teachers and three validated survey tools: the Creativity Fostering Teacher Behaviour Index, the Creativity in the Science Classroom Questionnaire, and the Culturally Responsive Teaching Self-Efficacy Scale. Key findings reveal that teachers consistently identified strong student-teacher relationships as fundamental to creating classroom environments where creativity flourishes. Understanding students' cultural backgrounds emerged as vital for building effective relationships, with trusting connections enabling students to take creative risks without fear of negative consequences. Teachers prioritizing relationship-building demonstrated greater willingness to adapt teaching methods to align with student interests, even when deviating from mandated curricula. The study provides evidence supporting educational methods that focus on both human connections and academic success, offering insights for teacher training, professional development, and education policy
Review of Use and Effectiveness of Health Impact Assessment in the Oil and Gas Sector
Get PDFHealth Impact Assessment (HIA) refers to a combination of procedures, methods and tools that systematically judges the potential, and sometimes unintended effects of a policy, plan, program or project on the health of a population and the distribution of those effects within the population. HIA identifies appropriate actions to manage those effects. HIA can be a useful tool to help the energy sector evaluate their health impacts in order to define their responsibilities to local communities and host governments. HIA identifies appropriate actions to manage those effects. Evidently, there is insufficient research at present to understand how HIA works in practice and if it is effective. Prior publications focused primarily on the public sector, general guidance and specific elements of HIA process. The overall goal of this research is to advance the state of knowledge of practice of HIA in the oil and gas sector in the context of current environmental and public health challenges by using both quantitative and qualitative approaches. The specific aims of this research are to determine whether and to what extent HIA process or practice varies from company to company within the oil and gas sector; to define HIA effectiveness within the oil and gas sector; and to illustrate whether the HIA process/practice is effective and under which circumstances. Existing HIA practices by oil and gas companies were surveyed via online questionnaire as part of sequential explanatory mixed methods (phase 1). To further illuminate HIA practice in contextual surroundings of companies which responded to the survey, several case studies were prepared (phase 2). This research seeks to understand the main drivers and barriers of an effective HIA to inform better processes and practice, and ultimately further the effectiveness of HIA within the energy sector. The paper presents the results of both phases and highlights main observations and findings. The research conclusions communicated to the key stakeholders identified, including energy companies, will make them more aware of their own and others’ views and practices, and challenges and opportunities that exist. It will empower them to take action toward change and inform the planning processes and aid the identification of future priorities of industry groups particularly IPECA (formerly known as International Petroleum Industry Environmental Conservation Association) and the International Association of Oil and Gas Producers (IOGP). It is anticipated that the recommendations could be applicable to the private sector companies in other industries
NON-TARIFF PROVISIONS IN MEGA-REGIONAL TRADE AGREEMENTS AND BACKWARD GLOBAL VALUE CHAIN TRADE IN THE AUTOMOTIVE INDUSTRY: AN EMPIRICAL ANALYSIS
No full textThis study empirically examines the impact of non-tariff provisions, including rules of origin (ROO), trade facilitation, and technical barriers to trade (TBT), in MegaRegional Trade Agreements (MRTAs), such as the Comprehensive and Progressive Agreement for Transpacific Partnership (CPTPP) and the United States-MexicoCanada Agreement (USMCA), on member countries’ participation in automotive Global value chains (GVCs). Using a combination of quantitative methods—including Chow test, gravity model, and natural language processing (NLP) analysis—and qualitative analysis, the research finds that deeper MRTA provisions are associated with increased backward GVC trade, especially in regions with robust manufacturing bases.
The findings highlight that while global crises and policy shifts have slowed GVC growth since 2008, MRTAs can play a crucial role in revitalizing cross-border production networks. The study provides actionable insights for policymakers considering MRTA accession and for designing trade agreements that enhance industrial competitiveness and resilience
The effect of low-cost food storage hardware (meatsafe) and minimal behavior change communication (BCC) on microbiological contamination of complementary foods, and complementary food feeding practices by caregivers of children under 2 in urban slum households in Dhaka, Bangladesh
Get PDFFood hygiene remains under-addressed in water, sanitation, and hygiene (WASH) interventions, even though nearly half of complementary foods in urban Bangladesh harbor fecal bacteria and uncovered dishes have twice the contamination prevalence. This dissertation examined whether a low-cost storage device could reduce microbial contamination of children’s food and diarrhea in Dhaka’s informal settlements.
We embedded a randomized controlled trial within Korail, Dhaka’s largest slum, enrolling 290 households with children aged 6–24 months. Households were randomized 1:1 to receive either a wire-mesh food cabinet (“meatsafe”) plus two pictorial behavior-change posters on food hygiene, or usual care. The ventilated cabinet was intended to shield food from flies, dust, and animals while allowing steam to escape. Complementary-food samples and caregiver-reported 7-day diarrhea were measured at baseline and five fortnightly visits over twelve weeks. A purposive subsample of 15 intervention caregivers completed in-depth interviews exploring barriers and facilitators to correct use. Secondary analyses assessed passive effects on commercial-snack provision and WHO Infant and Young Child Feeding indicators.
Uptake was high: 88–99 % of intervention households stored cooked food in the meatsafe at every visit, and fully covered dishes rose from 67% to 88%. Yet the intervention failed to lower E. coli ≥100 CFU/g (PR = 0.85, 95% CI 0.56–1.27) or caregiver-reported diarrhea (PR = 1.09, 95% CI 0.88–1.35). Snack consumption (95–99% at baseline) and dietary-diversity indicators also remained unchanged. Qualitative findings illuminated critical behavioral and contextual gaps: nearly all caregivers delayed storage to avoid “steam spoilage” and melting shelf liners, leaving freshly cooked food exposed during the key contamination window. Chronic fuel scarcity (7–8 families sharing one stove), seasonal flooding, and pest intrusion through mesh gaps further eroded potential benefits.
Results echo other large-scale WASH trials: single-purpose tools and minimal messaging seldom offset systemic microbial risks rooted in poverty, crowding, and fragile infrastructure. Effective food-safety strategies will require redesigned hardware aligned with local routines, sustained behavior-change engagement, and structural investments in reliable energy, waste services, and pest control
FROM COMPLIANCE TO COLLABORATION: BUILDING TRUST AND AUTONOMY FOR EFFECTIVE FEEDBACK
No full textThis study investigated the relationship between K-2
teachers’ trust in their principals, perceptions of professional
autonomy, and receptiveness to instructional feedback in East
Coast urban Title I elementary schools. Grounded in
Bronfenbrenner’s ecological systems theory, this study explored
how these factors intersect to influence instructional conditions
affecting Black student achievement. A convergent parallel mixed
methods design was used. Quantitative data from 28 teachers were
collected using the Omnibus Trust Scale and the Teacher Autonomy
Scale. Feedback receptiveness was measured using semistructured interviews with six teachers.
Regression analysis revealed that trust was significantly
predicted by all four trust domains: benevolence (β = 0.37, p =
<0.01), reliability (β = 0.33, p = <0.01), competence (β = 0.27, p =
<0.01), and honesty (β = 0.13, p = <0.09). The overall model was
significant F (4,23) = 28.09, p = <.001, and explained 87% of the
variation in trust (R2 = 0.87). Both teaching autonomy (β = 0.74, p =
<.001) and curriculum autonomy (β = 0.44, p = <.001) were
statistically significant predictors of overall autonomy (F (5,25) =
2.29, p = <.001, R2 = 0.97), with the model accounting for 97% of
the variance (R2 = 0.97). Interview data reinforced the importance of leader reliability, benevolence, and competence in shaping
feedback receptiveness.
These findings informed the development of The Feedback
Partnership, a podcast-based professional learning intervention,
grounded in experiential learning and designed to foster trust and
enhance feedback receptiveness in urban, Title I K-2 settings
UNDERSTANDING ALIGNMENT FOR VISION AND LANGUAGE MODELS
No full textAlignment plays a central role in enabling vision and language models (VLMs) to generalize, adapt, and perform effectively across a wide range of tasks and domains. This thesis investigates alignment as a key principle across multiple stages of model development, providing a unified perspective on how it supports robust and flexible multimodal learning. The study begins by examining how vision models can be adapted across diverse domains, demonstrating that aligning visual representations helps improve generalization in novel environments. Experimental results show that well-aligned feature spaces significantly reduce performance gaps caused by domain shifts. Building on this, the focus shifts to aligning vision models with language models for open-vocabulary object detection. In this setting, alignment enables the recognition of categories beyond the fixed label set used during training. Methods are proposed to bridge the gap between textual descriptions and visual features, allowing for more scalable and flexible object recognition. The work further explores panoptic segmentation by aligning segmentation foundation models with language supervision, where the language guidance helps enable panoptic segmentation in an open-vocabulary setting, leading to more comprehensive and context-aware scene understanding. Finally, the role of alignment is studied in large multimodal language models, with particular attention to hallucinations cases where the model produces incorrect or fabricated outputs. A hallucination score is introduced based on consistency checks to quantify fine-grained hallucinations and assess their severity. Together, these contributions present a comprehensive investigation of alignment in vision and language models, highlighting its importance for building adaptable, scalable, and reliable multimodal systems