19304 research outputs found
Sort by
Black Emerging Adults And Law Enforcement: Perceptions and Experiences
No full textObjective: Emerging adults – particularly Black emerging adults—are at a heightened risk of law enforcement encounters compared to other racial groups. This may be attributed to the intersecting identities of Black and emerging adulthood. The present study investigated the current perceptions and experiences held by Black emerging adults regarding interactions with law enforcement. Methods: Informed by Bronfenbrenner’s Ecological Systems Theory (1979), we conducted individual semi-structured with six Black emerging adults with a history of encounters with law enforcement. Interviews were transcribed and subjected to a reflexive thematic analysis. Results: Through a reflexive thematic analysis, five key themes were constructed: Emerging Adulthood as the “In-Between” Phase of Life, Mixed Perceptions of Law Enforcement, Direct Experiences with Law Enforcement, Law Enforcement Improvements and the Black Community and Improvements to Law Enforcement Training. Aligned with Bronfenbrenner’s model, Black emerging adults formed perceptions of law enforcement based on environmental factors such as experiences of others as well as direct experiences. Conclusion: The present study adds to a growing body of literature that examines the relationships between Black emerging adults and law enforcement from a qualitative lens
Pluralizing Conscience: A Critical Assessment of the Liberal Theory of Conscience in Twentieth-Century Thought
No full textThe idea of conscience held a prominent place in the history of Western moral philosophy before falling into disrepute in the early twentieth century. New theories put forward by sociologists and psychoanalysts, and advanced by twentieth century liberalism, emphasized the descriptive and developmental experience of conscience over metaphysical and theological speculations about its fundamental nature. Because the experience of moral obligation and motivation—with which conscience is often associated—is common to people of different faiths and creeds, conscience became detached from comprehensive doctrines that determined features like its origin and reliability. As a result, for many liberal thinkers, conscience holds the unique promise of being a politically neutral and pluralistic concept, something which can be assented to by all as that which imbues human beings with their inherent dignity and worth. This dissertation argues that conscience is not, and has never been, politically neutral. All conceptualizations of conscience—even the vague sense of conscience as sincere private conviction, which twentieth century liberalism was keen to promote—presuppose certain ideas about human nature, in both what we can know and how we come to know it. As such, those who believe that conscience can be neutral have, in effect, moved the focus from 'the good' itself to that-by-which we come to know 'the good.' This points to a fundamental irony in liberal thought: that freedom of conscience can moderate competing concepts of ‘the good,’ but not competing concepts of conscience itself. This dissertation serves to point out this irony, which continues to be overlooked by most of the twentieth- and twenty-first-century scholarship on conscience. In doing so, it pursues the important and necessary tasks of returning to conscience its rightful complexity and nuance in moral-political discourse, and of restoring faith in democratic institutions
“Everything Sounds Great on Paper”: Drivers of Housing Instability as Youth Transition out of Foster Care
No full textTransition-age youth exiting foster care (TAY) are at high risk for housing instability, with nearly half experiencing homelessness before age 26. Multi-level factors are associated with greater risk, including individual, social, and geographic contexts. This study explored experiences of TAY in a large region of Texas to identify drivers of housing stability during the transition out of care. Youth aged 18–25 who were connected to the region’s foster care transition center were recruited to participate in a mixed-methods, semi-structured interview (n = 25). Youth were prompted to identify networks of up to 20 people who had provided support over the past year. Interview questions explored what happened when youth turned 18, including changes in their housing situations, and delved into relationships with the network members. An iterative coding process was used to create a matrix to examine housing transitions and social supports within and across cases, then identify themes and subthemes. Housing instability was common, with 13 of 25 participants reporting episodes of homelessness after turning 18. Abrupt transitions were driven by systemic factors related to placement settings, strict rules, and a lack of available housing options. Social network data illuminated the close link between housing and the social network, along with the importance of “housing-capable” adults who helped prevent homelessness. Findings call for the development of more youth-friendly housing options for TAY transitioning out of care and interventions that help to build enduring social supports
Enhancing the Robustness in Reinforcement Learning: Domain Knowledge Integration, Uncertainty Quantification, and Risk-Sensitive Strategies
No full textReinforcement learning (RL) has become a powerful framework for sequential decision-making, particularly in complex environments where traditional optimization methods struggle. This dissertation develops novel RL-based methodologies that address key challenges in dynamic resource allocation, statistical inference, and risk-sensitive decision-making. The contributions span three major directions and emphasize both theoretical rigor and practical relevance. The first part introduces a domain knowledge-informed reinforcement learning (RL) framework for dynamic resource matching in manufacturing. The problem is modeled as a Markov Decision Process (MDP), where infeasibility and prior-policy-based penalties are integrated into tabular Q-learning and extended to the deep deterministic policy gradient (DDPG) algorithm. This yields the domain knowledge-informed DDPG (DKDDPG) framework, which is designed to address high-dimensional, continuous decision spaces. We establish theoretical convergence guarantees for the tabular case, and empirical results demonstrate improved learning efficiency and solution quality in both small- and large-scale instances. The second part focuses on the statistical inference properties of RL algorithms. We propose a generalized sample-averaged Q-learning algorithm that aggregates rewards and transitions over time-varying batches. Leveraging the functional central limit theorem (FCLT), we establish asymptotic normality for the value estimates and introduce a random scaling technique for interval estimation that avoids explicit variance estimation or hyperparameter tuning. Simulation results demonstrate improved inference accuracy and coverage control under various batch scheduling strategies. The third part presents a variance-penalized RL framework that integrates empirical variance estimates into both Q-learning and actor-critic updates. Using nonparametric tools like online bootstrapping and random scaling, we enable risk-sensitive learning without auxiliary critics. We provide convergence analysis under standard assumptions and empirically show enhanced policy stability in stochastic settings. Together, these contributions offer a unified framework for integrating domain knowledge, uncertainty quantification, and risk sensitivity in reinforcement learning. The results provide both theoretical insight and practical tools for advancing decision-making in complex, uncertain environments
Prediction/Assessment of CO2 EOR and Storage Efficiency in Residual Oil Zones Using Machine Learning Techniques
No full textResidual oil zones (ROZ) arise under the oil&ndash;water contact of main pay zones due to diverse geological conditions. Historically, these zones were considered economically unviable for development with conventional recovery methods because of the immobile nature of the oil. However, they represent a substantial subsurface volume with strong potential for CO<sub>2</sub> sequestration and storage. Despite this potential, effective techniques for assessing CO<sub>2</sub>-EOR performance coupled with CCUS in ROZs remain limited. To address this gap, this study introduces a machine learning framework that employs artificial neural network (ANN) models trained on data generated from a large number of reservoir simulations (300 cases produced using Latin Hypercube Sampling across nine geological and operational parameters). The dataset was divided into training and testing subsets to ensure generalization, with key input variables including reservoir properties (thickness, permeability, porosity, Sorg, salinity) and operational parameters (producer BHP and CO<sub>2</sub> injection rate). The objective was to forecast CO<sub>2</sub> storage capacity and oil recovery potential, thereby reducing reliance on time-consuming and costly reservoir simulations. The developed ANN models achieved high predictive accuracy, with R<sup>2</sup> values ranging from 0.90 to 0.98 and mean absolute percentage error (MAPRE) consistently below 10%. Validation against real ROZ field data demonstrated strong agreement, confirming model reliability. Beyond prediction, the workflow also provided insights for reservoir management: optimization results indicated that maintaining a producer BHP of approximately 1250 psi and a CO<sub>2</sub> injection rate of 14&ndash;16 MMSCF/D offered the best balance between enhanced oil recovery and stable storage efficiency. In summary, the integrated combination of reservoir simulation and machine learning provides a fast, technically robust, and cost-effective tool for evaluating CO<sub>2</sub>-EOR and CCUS performance in ROZs. The demonstrated accuracy, scalability, and optimization capability make the proposed ANN workflow well-suited for both rapid screening and field-scale applications
The Current Use and Expansion of “The New Feminist Absurd”: An Investigation Into Why Twenty-First Century Female Artists Are Sharing Their Voices Through This Genre
No full textThis thesis defines how works from the past ten years fit within and even expand on the trend of the “New Feminist Absurd,” as defined by scholar Emily B. Klein. I present three twenty-first century plays that showcase the tenets of the “New Feminist Absurd”: Ruby Rae Spiegel’s 2015 Dry Land, Rotem Nachmany’s 2019 Scrambled, and Anikka Lekven and Molly Fonseca’s 2023 Woman Managing Her Anxiety. I analyze these shows through script analysis and performance analysis with a focus on semiotics due to the absurdist material. I also share interview material, both published and personal, in order to investigate why current-day feminist artists are drawn to the genre of the “New Feminist Absurd”. With these three plays, one can see how Klein’s criteria of the genre present themselves within an early work of the genre and how they are being expanded in the genre’s newest works, allowing for predictions of its future. This thesis demonstrates that the playwrights for each of these plays found themselves drawn to this nontraditional genre since they felt that the realism typically found on stages could not fully express their stories or emotions, which speaks to why combining feminist ideas with absurdist theatre is more popular today than ever before
Artificial Intelligence and Automation and Its Effect on Foreign Labor
No full textStudies artificial intelligence's (AI) impact on skilled and unskilled immigrant labor in the United States (U.S.). By utilizing other studies, papers, graphs, and data the impact of AI on foreign born workers can be illustrated. Based on the findings and trends, we discussed policy reformations and laws. These policy ideas will help lessen the friction between AI and foreign-born workers.Economics, Department ofHonors Colleg
The Role of Hyaluronan in the Ocular Surface
No full textPurpose: The extracellular matrix (ECM) has a vital role in all tissues of the body, including maintaining ocular surface homeostasis. Hyaluronan (HA) is found in the extracellular matrices of the corneal limbus and meibomian gland (MG). This dissertation evaluated the role of hyaluronan in ocular tissues, particularly in the limbal stem cell niche and the meibomian gland. Methods: (1) To determine the role of HA in the limbal stem cell niche, primary human limbal epithelial stem/progenitor cells were isolated and expanded onto differently coated cultures with or without HA and the yield of limbal epithelial stem cells assessed. (2) To characterize the limbal stem cells niche, the ECM of the mouse corneal buttons and limbal rims was analyzed by various proteomics approaches. (3) To assess the role of HA in age-related meibomian gland dysfunction (ARMGD), Has1-/-;Has3-/- (hyaluronan synthase 1 and 3) knockout mouse Meibomian glands (MGs) were examined as they aged and glandular area, dropout, and atrophy assessed. (4) In order to provide a more translational model of ARMGD to humans, non-human primate (NHP) eyelids at varying ages were assessed and compared to human eyelids. Results: (1) Limbal stem cells cultured onto HA retained their stem cell characteristics such as putative stem cell markers, cell size and roundness, holoclone colony formation capability, and showed decreased senescence over time compared to other coatings. (2) There are differentially expressed ECM genes between the central cornea and limbal region, including higher expression of HA binding proteins TSG-6 and IaI heavy chains in the limbal region. (3) Has1-/-;Has3-/- knockout mice did not show signs of ARMGD at 1yr. of age and minor gland atrophy and dropout at 2yrs. of age in comparison to the wild type mice. (4) The non-human primate presented with symptoms of ARMGD as they aged and present with gland tortuosity and hooking in the lower eyelids. Conclusions: HA was able to maintain limbal stem cell stemness during ex vivo expansion and a specialized HA-rich stem cell niche was found in the limbus. There is also a HA matrix surrounding the MGs of mice which prevented ARMGD. NHPs develop ARMGD similar to humans
The Transition to Kindergarten for Autistic Children: A Three-Part Study Examining Systems Level Influences on the Transition
No full textBackground: For children on the autism spectrum, the transition into kindergarten presents unique challenges that may require specific supports and services. Despite research highlighting effective ways to help prepare for and begin kindergarten, caregivers of children with this diagnosis report challenges receiving adequate support and services for their child during this transition. Moreover, socioeconomically, culturally, and linguistically diverse (SCLD) caregivers have reported additional challenges during the transition. Purpose: Broadly, the goal of these studies is to examine parents’ and teachers’ perspectives on and experiences during the transition to kindergarten. Study 1 explored pre-service teachers’ knowledge and attitudes about autism and the transition to kindergarten for autistic children and examined predictors of the importance of transition practices. Study 2 aimed to compare the transition experiences of English and Spanish-speaking caregivers of autistic children, before and during the Covid-19 pandemic. Study 3 used focus groups to explore in-depth and compare the experiences of Hispanic/Latine caregivers of children with autism who transitioned to kindergarten during the Covid-19 pandemic. Methods: Study 1 collected data from 71 pre-service teachers from across the United States. Study 2 used a survey to collect data from 200 SCLD caregivers of autistic children who were in kindergarten at some point within the past five years. MANOVA analyses examined the association between time of transition (pre vs. during the pandemic), race, ethnicity, generational status, and language with the use of transition practices and perceived transition success. Study 3 used a thematic analysis (Braun & Clarke, 2003) to analyze qualitative data collected from four Hispanic/Latine mothers of autistic children about their child’s kindergarten transition experience. Results: A multiple regression analysis revealed knowledge of autism, knowledge of transition practices, and beliefs about how others perceive the transition to kindergarten did not predict ratings of importance of transition practices. MANOVA analyses revealed that special education pre-service teachers had higher knowledge of autism and subjective norm belief scores compared to other pre-service teachers (Jellinek et al., 2022). In Study 2, findings indicated SCLD caregivers experienced challenges during their child’s transition including their child’s behavioral and social difficulties, school and teacher knowledge, and communication and relationship difficulties with their child’s school. Moreover, there was significantly lower use of transition practices and rated transition success for first generation immigrants compared to third generation immigrants, and for Black and White participants who identified as Hispanic/Latine compared to those who do not. In Study 3, findings highlighted that Hispanic/Latine mothers reported kindergarten transition challenges including developing and maintaining trusting relationships with their child’s school, a lack of knowledge about autism by teachers and within their community, and that language and communication barriers influenced their experiences and their child’s during the kindergarten transition. Further, parents reported a commitment to advocating for their child and distinct facilitators including talking with other parents and trusted providers. Discussion: The transition to kindergarten is a pivotal process as it lays the foundation for a child’s future schooling, yet there are many barriers to success that children with autism encounter during this process. Outcomes from these studies will further clarify barriers to the transition process and present opportunities to bolster the systems that support marginalized autistic children and their families during this process
Characterization of Novel Liver X Receptor Inverse Agonist in Breast Cancers
No full textBreast cancer is the second most diagnosed cancer type in women and is responsible for approximately 13% of cancer-related mortalities. Significant advances have been made in breast cancer therapy to date. However, the development of targeted therapies remains essential to overcome the lack of treatment options for TNBCs and acquired resistance. Metabolic reprogramming, wherein cancer cells rewire their metabolism for growth and survival, is a critical area of interest. Developing anti-metabolic drugs is an emerging field that targets the upregulated metabolic dependence of cancer. The liver X receptors (LXR) are ligand-modulated nuclear receptors. LXR activation upregulates the expression of genes involved in altered metabolic pathways in cancer, including cholesterol transport, glucose, and lipid metabolism. Moreover, LXR inhibition using the inverse agonist SR9243 has been shown to disrupt glycolysis, de novo lipogenesis, and induce antitumor immunity. In our lab, we discovered another cancer-specific LXR inverse agonist, GAC0001E5 (1E5), in pancreatic ductal adenocarcinoma (PDAC). Treatment with 1E5 has been demonstrated to inhibit glutaminolysis and induce oxidative stress. Additionally, LXRβ transcript levels are upregulated in breast cancer patient tissue. Therefore, we posit that characterization of the effects of 1E5 in breast cancer can reveal potential antitumor effects. Compared to LXR agonist GW3965, inverse agonist 1E5 treatment markedly disrupted breast cancer cell proliferation by downregulating LXR target genes and inducing LXRβ protein degradation. Mechanistic evaluations of 1E5 revealed that it downregulates glutamine metabolism in breast cancers. Specifically, 1E5 downregulated glutaminase 1 (GLS1) gene expression, leading to decreased levels of intracellular glutamate and glutathione, which elevated oxidative stress due to the accumulation of reactive oxygen species (ROS). Moreover, targeting FASN, a de novo lipogenesis enzyme, is an emerging target in HER2-positive breast cancer. Treatments with 1E5 reduced FASN and HER2 protein levels. Here, we suggest that there is a potential crosstalk between LXR, FASN, and HER2, and that inhibition by 1E5 can downregulate HER2-mediated oncogenic signaling and induce apoptosis. Taken together, these findings lay the foundation for the use of pharmacological LXR inhibition as a potential therapeutic strategy in targeting various aspects of breast cancers