Now showing items 1-20 of 4159


      Surdeanu, Mihai; Leon, Daniel Ixim (The University of Arizona., 2024)
      This paper explores the challenges of hallucination and misinformation spread by large language models (LLMs), focusing on the importance of their accuracy and reliability for use in health and inclement weather information dissemination. To address these issues, we introduce the AZX chatbot, which utilizes a retrieval-augmented generation (RAG) process. This method involves compiling trusted sources into a dense passage retrieval database of vectorized text documents to support informed response generation. The chatbot is designed to utilize these sources in providing its response in order to provide users with transparency regarding its logic and enable them to verify its responses. This chatbot, alongside ChatGPT and Gemini, was assessed in its responses to various prompts related to shelter, inclement weather alerts, health hazards, and disease. It slightly outperformed these baselines for prompts related to shelter and health hazards, but only succeeded on about half of the prompts in these categories, indicating the need for further improvement. However, it excelled and far outperformed the baselines for prompts related to weather and disease. Further investigation should look into improving the formerly described areas, as well as methods for efficiently updating the sources within the database.

      Wilson, Amanda; Lasher, Joshua Keith (The University of Arizona., 2024)
      Introduction: Access to clean drinking water is a fundamental human right for sustaining health and well-being. However, unhoused individuals face significant barriers to accessing this essential resource, particularly in regions like Tucson, Arizona, where access to clean drinking water and public resources is limited. This study aimed to address this gap by examining the experiences of unhoused individuals regarding their access to clean drinking water in Tucson. Methods: Unhoused individuals (n=29) were surveyed between November 1st and December 13th, 2023, with the assistance of El Rio Health to oversee the surveys. This study protocol was approved by the University of Arizona Institutional Review Board. The surveys consisted of a structured questionnaire that explored various aspects of water access, including sources, challenges, seasonal variations, storage methods, and perceived difficulty accessing and storing clean drinking water. Results: Survey responses reported that convenience and grocery stores were common drinking water sources for unhoused individuals in Tucson. Denied access and financial constraints were identified as primary challenges. Seasonal variations in water availability were noted, particularly during the summer months, when water availability is more difficult to obtain reliably. Reusable containers and single-use plastics were the most prevalent storage containers, with varying cleaning practices reported. Discussion: The study's findings emphasize the urgent need to address water insecurity among unhoused populations. They emphasize the importance of equitable access to clean drinking water for promoting health and social justice. Legal implications include the need for policies mandating water provision in public spaces. Educational interventions could empower unhoused individuals with knowledge about the importance of hydration. While the study's limitations, such as sample size and potential bias, warrant consideration, its insights have broader applicability beyond Tucson, informing interventions in urban settings globally. Overall, this study serves as a catalyst for collaborative efforts to ensure equitable access to clean water and enhance the well-being of unhoused communities.

      Wilson, Dalziel; Land, Andrew Thomas (The University of Arizona., 2024)
      Material absorption is a key limitation in nanophotonic systems; however, its characterization is often obscured by scattering and diffraction loss. Here we show that nanomechanical frequency spectroscopy can be used to characterize the absorption of a dielectric thin film at the parts-per-million (ppm) level, and use it to characterize the absorption of stoichiometric silicon nitride (Si3N4), a ubiquitous low-loss optomechanical material. Specifically, we track the frequency shift of a high-Q Si3N4 trampoline resonator in response to photothermal heating by a ~ 10 mW laser beam, and infer the absorption of the thin film from a model including thermal stress relaxation and both radiative and conductive heat transfer. A key insight is the presence of two thermalization timescales, a rapid (~ 0.1 sec) timescale due to radiative thermalization of the Si3N4 thin film, and a slow (~ 100 sec) timescale due to parasitic heating of the Si device chip. We infer the extinction coefficient of Si3N4 to be ~ 0.1 - 1 ppm in the 532 - 1550 nm wavelength range, comparable to bounds set by waveguide resonators and notably lower than estimates with membrane-in-the-middle cavity optomechanical systems. Our approach is applicable to a broad variety of nanophotonic materials and may offer new insights into their potential.

      Andrews-Hanna, Jessica; Koch-Kreher, Lucas (The University of Arizona., 2024)
      In this thesis, the relationship between emotional granularity, empathy, and relationship satisfaction was examined. Emotional granularity is defined as the ability to make subtle distinctions between emotional states. While a more novel concept, a growing body of literature links lower granularity to a variety of psychopathologies and generally lower psychological well-being. In contrast, higher granularity has been found to be associated with better psychological outcomes, yet less research has been done in the context of relationships and even less with its relation to empathy. Initially, a mediation relationship was considered with empathy being the mediator between emotional granularity and relationship satisfaction. Subsequently, an interaction between emotional and granularity was modeled as the mediation did not show significant results. The interaction models yielded significant results with emotional granularity being a significant predictor of lower relationship satisfaction given lower empathy, however higher empathy counteracted this effect. Additionally, emotional granularity is quantified in this study utilizing text analysis methodology, which deviates from prior literature, but provides insightful looks into different approaches for measuring the construct.

      Blume, Andreas; Kinnebrew, Briana SantaCruz (The University of Arizona., 2024)
      This thesis considers the effect of object sequencing within art auctions on revenue. Prominent auction houses such as Sotheby's and Christie's generate billions of revenue supported by a multitude of professional decisions. The sequence objects are sold in is worthy of discussion. This thesis presents two models that utilize second-price auction formats and consider if and when object sequencing matters to revenue.

      Barr, Sandra; Kinnebrew, Briana SantaCruz (The University of Arizona., 2024)
      Within Art History many women artists have been left out of intellectual discussion. Many women artists were often systematically excluded from receiving training or work. There are many female artists who produced works during every major art historical movement, who still aren't researched or examined. There are many female artists who produced works during every major art historical movement, who still aren't researched or examined. This thesis seeks to contribute to further academic inquiry of women artists and their self-portraits. Many female artists were only accepted as portraitist, thus women artists who complete self-portraits are faced with the dual conundrum of representing themselves aesthetically and presenting their work as substantially impressive for a discipline that may not want them there. This thesis focuses on three female artists, Artemisia Gentileschi, a 17th century Baroque artist; Mary Cassatt, a 19th century Impressionist; and Remedios Varo, a 20th century Surrealist. There is a discussion of their careers and an analysis of one of each of their self-portraits. This thesis contains a final section about my experience creating my own self-portrait.

      Gothard, Katalin; Kim, Sun Woo (The University of Arizona., 2024)
      Adolescence marks a pivotal period of development in brain structure and behavior. The changes in how our brains are wired directly influence how we respond to the world around us. The uncinate fasciculus (UF), which develops during adolescence, is a major white matter tract that connects the orbitofrontal cortex (OFC) and the amygdala. The OFC and amygdala are heavily involved in reward processing and social decision-making. To investigate how brain maturation influences these behaviors, we collected diffusion MR images and behavioral measurements (a non-social delay discounting task and a social reciprocation task) in four adolescent rhesus macaques (31 to 51 months old). Two MR images were taken about a year apart to detect any changes in diffusion indices such as fractional anisotropy (FA) and mean diffusivity (MD), which can inform us about myelination. We first analyzed whether these diffusion indices change over age, and then whether they predict behaviors in the two tasks during adolescence. We found that MD in the OFC and UF followed an inverted U curve across age, indicating ongoing brain development through adolescence. MD in the amygdala and FA in all three brain regions did not follow the same pattern across age. High MD was correlated with a higher willingness to wait for a reward in the non-social delay discounting task. Higher FA in the OFC and amygdala showed a correlation with higher prosocial decisions in the social reciprocation task. The changes in MD across age suggest naturalistic synaptic pruning followed by strengthening of myelination. A higher MD, associated with a decrease in myelination, predicts a higher tolerance for delayed reward. Higher FA, reflecting higher axonal integrity, predicts more prosocial decisions. This may be due to the strengthening of axons within the OFC and amygdala supporting the reward valuation of others, not only for themselves. Overall, we found certain anatomical features measured by diffusion MR images show changes with age and support both reward valuation and social decisions.

      Goldman, Steven; Kazui, Saki (The University of Arizona., 2024)
      Cardiac arrhythmias often present with complex set of symptoms and etiologies, and clinicians are tasked with determining the nature of arrhythmias and developing a therapy tailored to individual needs. Notably, atrial fibrillation (AFib) is the most common arrhythmia in adults in the United States. Atrial fibrillation results in desynchrony of electrical activity in the atria leading to abnormal muscle contractions and increasing the risk of stroke. Uncovering some of the potential cellular mechanisms and the risk factors that promote the progression of AFib is important. Ventricular tachycardia (VT) is another type of arrhythmia with a myriad of etiologies encompassing a wide range of symptoms. It is a major contributor to the onset of sudden cardiac death (SCD) in patients, and multiple therapies have been developed to manage the disease. Furthermore, the use of cell-based therapy in the form of a biologic platform has emerged as an intriguing new approach to treating heart diseases. In this review, the characteristics and etiologies of AFib and VT are explored, along with their current therapies and the potential role of an active biologic platform in future treatments of heart disease.

      Gonzalez, Frank; Karlapudi, Pragnya Varna (The University of Arizona., 2024)
      Dobbs v. Jackson may impact residency program applications because of restricted abortion care. Restricted abortion care has led to a limited scope of practice within physicians who practice OB/GYN care. This has bled into residency program curriculum, indicating that applications to certain schools may go up or down. Medical students were surveyed with a 2-condition experiment to see their program preferences. The experimental group was primed on information regarding Dobbs v. Jackson before answering questions about their preferences for residency programs. The control group was not primed with information prior to taking the survey. The rest of the survey then asked about medical students' residency program preferences. Overall, medical students had significantly higher preferences for Democrat-dominated states. However, there was no evidence of effect from the conditions. The priming did not reveal any significant differences in residency program preference. It would be interesting to gather more data to see if there is evidence of effect. Dobbs v. Jackson is impacting the future of healthcare and may lead to larger disparities in abortion care.

      Limesand, Sean; Johnson, Eliza Hazel (The University of Arizona., 2024)
      Prior research indicates that constructs containing the endogenous ovine insulin gene promoter have little to no activity in pancreatic beta cells. Our objective was to develop an ovine insulin promoter that is both glucose-responsive and beta-cell specific. To do this, we concatemerized the proximal promoter region of the ovine insulin gene to create a sheep insulin super promoter (SISP). We cloned SISP into two adenovirus constructs with a CMV promoter driving the fluorescent protein ZsGreen and SISP promoting either the fluorescent protein mCherry or luminescent protein luciferase. These constructs were tested in MIN6 cells to determine the promoter's specificity and glucose responsiveness in an insulinoma cell line. Results indicate that SISP had a high infection efficiency in MIN6 and was responsive to higher glucose concentrations. The addition of the antioxidant N-acetyl cysteine had no effect on promoter responsiveness. The next steps for this super promoter include testing in ovine islets and directly comparing its activity to the endogenous ovine insulin promoter. SISP can be used as a tool for beta cell isolation, monitoring glucose-stimulated insulin secretion, and targeting gene expression in beta cells.

      Mitchell, Rachel; Javier, Virginia Gabrielle (The University of Arizona., 2024)
      Stinknet (Oncosiphon pilulifer) is a winter annual forb and an emerging invasive plant species in the American southwest. Over the past 50 years, populations have established in several counties of California and Arizona. Stinknet spreads rapidly, increasing the continuity of vegetation in arid landscapes, therefore increasing fire risk and emission of VOCs dangerous to human health. Despite its noxious weed status in multiple counties, little is known about the basic biology of stinknet. Here, we collected seeds from four populations of stinknet located in San Diego County, CA, Maricopa County, AZ and Pima County, AZ. We quantified germination rates under several conditions and examined the response of these four populations to experimental drought in a greenhouse experiment. Seeds from four populations of stinknet were exposed to three germination temperatures and two different burial treatments during germination trials. We found a significant difference in the time-to-event curves in the preliminary germination experiment (p=0.005) and recorded only a an average rate of 9.575% germination rate from both experiments The results of our germination experiment suggest that stinknet seeds have low germination rates across all populations, regardless of seed source location, and do not require cold stratification to germinate. The same four stinknet populations also underwent a six-week greenhouse-based experiment to understand drought tolerance and plant functional trait expression under a twice-a-week (full) water treatment and a control daily watering (even) treatment. Plants were grown under drought conditions, but received consistent or pulsed water. The results of the greenhouse experiment suggests that watering treatment, as well as the mean annual precipitation and mean annual temperature of the seed collection site, influence height and growth rate of greenhouse grown plants. However, mean annual precipitation and temperature are the only significant influences on mortality and root:shoot suggesting that some populations are more susceptible to drought conditions regardless of whether water is pulsed or consistent. Understanding how this species responds to various germination, drought, and precipitation conditions will help land managers understand which vegetation communities are at the greatest risk of invasion and how to control these populations based on certain external factors like precipitation amount and frequency.

      Rozo, Eduardo; Hughes, Aaron Dean (The University of Arizona., 2024)
      This paper describes a method to compute the spectroscopic richness of galaxy clusters. This method stems from the redMaPPer photometric cluster finder and the idea that spectroscopic data is able to better discriminate between member and non-member galaxies. Similar to redMaPPer, the method outlined here defines a membership probability for each galaxy around a central galaxy of a cluster. However instead of utilizing the position, color, and magnitude information of the galaxy, the spectroscopic method utilizes the velocity information of galaxies around a cluster to determine if galaxies with a given position and velocity are likely to be cluster members or not. Running on halos from the Cardinal mock galaxy catalog and determining the richness of the cluster in each halo yields a mass-richness relationship. Comparing the mass-richness relation found from this spectroscopic method and the relation from redMaPPer yields a clear decrease in the scatter of 15 percent at low mass and 25 percent at high mass. This reveals the promise of incorporating a spectroscopic component into cluster finder algorithms.

      Ottusch, Timothy; Im, Anna ThakYa (The University of Arizona., 2024)
      For my Honors thesis, I chose to do something that is a bit more unique. I researched, wrote, and produced a podcast called Family Talk. Family talk aims to bridge the gap between academia and everyday people by educating the public on Family Systems theory. Family Systems theory proposes that the family is dynamic and interconnected, and that events that happen within and without the family impact the whole system.The first episode is about Family Systems Theory and how it is useful for understanding families. The second episode is about boundaries and communication in the family, the third episode is about mental health diagnoses in the family, and the final episode is about nontraditional family forms. Each episode defines Family Systems Theory concepts, overviews a case study, and then gives recommendations for improving family functioning. My goal was to make this something that would be accessible to families and as well as for professors to distribute to their students.

      Venkataramani, Sangeetha; Hsu, Kaitlyn De-Ping (The University of Arizona., 2024)
      For years, China has been at the forefront of development, innovation, and global interconnectivity. China's cultural history is both rich and ancient, illustrating a long-standing trajectory of cultural advancement and transformation. This cultural transformation is marked by significant changes, as new generations introduce and adopt distinct preferences, thereby reshaping the cultural and economic landscape. Currently, the international community is beginning to more profoundly appreciate and experience the influence of China's robust economy and its dynamic consumer markets. Partners at McKinsey & Company say, "In a fast-changing market such as China's, finding new ways to serve consumers at all income levels is an imperative--and leading Chinese players are creating the proof of concept for many other global companies" (2021). Hence, in this thesis, I will explore and answer the following question: How have cultural factors in China shifted in the past decade and influenced consumer behavior in the marketplace? Furthermore, I will address how marketers can effectively tailor their marketing strategies to accommodate and align with these nuances. The subsequent sections will consist of a thorough review of the literature and research, a comprehensive analysis of data, and informed recommendations. I chose to pursue this topic due to a strong interest in global marketing strategies and a deep desire to research a culture I identify with. I hope to apply the insights and knowledge gained from this study in future professional opportunities or life experiences that position me abroad.

      Stepanov, Misha; Hsu, Anthony Hui-An (The University of Arizona., 2024)
      The Event Horizon Telescope (EHT) has revolutionized our understanding of black holes by using Very Long Baseline Interferometry (VLBI) to reconstruct the highest resolution images of black holes. It combines data from a global network of radio telescopes to observe the immediate surroundings of black holes, particularly focusing on the supermassive black hole in the galaxy M87 and the center of our Milky Way. This Jupyter Book explores the intricate process of image reconstruction from EHT data, highlighting several critical aspects such as interpolation, loss calculation, and gradient computation methods. Interpolation schemes play a pivotal role in connecting the snap averaged data in EHT's sparse observation and recon- structing an image. By examining the relationship between the image and Fourier domains, we gain insights into the factors influencing high-resolution image recreation. One approach to image reconstruction involves creating a loss func- tion and minimizing it. The loss function quantifies how well the image matches the observational data and how same or expected the appearance of the image looks. The latter part of a loss function is called a regularizer. The optimum (the best performing image) is accepted as the image reconstruction of the data. Additionally, this notebook assesses two distinct methods for computing gradients: finite differences and an "dirtying the image" approach. By comparing their efficiency and accuracy, we aim to offer guidance on selecting appropriate gradient computation methods for image reconstruction tasks. Integrating these components, we perform gradient descent on sample test images to demonstrate practical applications of these theories.

      Harley, Heidi; Higginbotham, Enrico (The University of Arizona., 2024)
      Coordination in Tohono O'odham, a Southern Uto-Aztecan language spoken in Arizona and Sonora, Mexico, is sensitive to the syntactic categories of the conjuncts. Clausal coordination can be marked by an overt prefix ku- which is the synchronic descendent of the Proto Uto-Aztecan obviative subordinator *-ku ~ -ko and shares numerous cognates throughout the language family. The distribution of ku- in the left periphery is similar to the polar question marker n- and the subordinator m-. However, other typological characteristics of Tohono O'odham (e.g., auxiliary-second word order) closely interact with these elements in the left periphery of the clause. Drawing from corpus sources (e.g., Zepeda (1983), Mathiot (1973)) and previous literature on functional elements in Tohono O'odham (e.g., Hale (1983), Bare (2015)), this work aims to analyze the available instances of Tohono O'odham CP coordination within a Minimalist framework. Thus, it will present a clearer picture of Tohono O'odham's left periphery, permitting, in turn, more precise research questions in the future. It will also present new challenges to our understanding of the extended projection of the Tohono O'odham verb, opening the door to a cartography of O'odham functional elements.

      Smith, Steve; Hess, Erin (The University of Arizona., 2024)
      Stinknet is an invasive weed quickly invading the Southwest and Pima County. Due to its inedible nature, aggressive takeover of native plant cover and forage, and potential for fire fuel, I intended to investigate the effects of stinknet on small mammal populations and species composition. I set traps in stinknet infested areas and control sites in Tucson to test if there was a difference in the composition and abundance of small mammals. Though I caught nothing of interest I learned how to navigate land ownership, efficiently conduct trapping sessions, and gained valuable knowledge for future research in this topic.

      Goldsmith, Melissa; Herrmann, Carlie Maria (The University of Arizona., 2024)
      Synthetic oxytocin (OXT), also known as Pitocin, is commonly used in maternity care throughout the world, and is administered to women to induce or augment uterine contractions in labor. Due to frequent use of Pitocin in maternity care, it is critical to understand both the benefits and risks of Pitocin. Of interest, an emerging literature suggests that in addition to its positive benefits, Pitocin may also be correlated with disrupted breastfeeding patterns. The purpose of this study was to examine the relationship between total dosage of intrapartum Pitocin and breastfeeding patterns, defined as the amount of breastfeeding (%) vs formula feeding in the 3rd month after birth. In this study, no statistically significant correlations were found between intrapartum Pitocin dosage and subsequent breastfeeding patterns, but limited sample size and restricted range of data may have attenuated these findings. However, further statistical analysis comparing intrapartum Pitocin dosage level among high breastfeeding women at 3 months (80% or greater) with those breastfeeding less than 80% did show a statistically significant difference. Based upon these findings, further research is needed with a larger pool of participants to better understand and document the way in which intrapartum Pitocin influences breastfeeding.

      Ellingson, Katherine; Henert, Rianna Shay (The University of Arizona., 2024)
      Background: Type 2 diabetes, a non-communicable disease, affects hundreds of millions of people worldwide and is projected to continue increasing. Similarly, the global epidemic of antibiotic resistance from overuse is projected to increase as well. Recent studies have begun to establish evidence for an association between excessive antibiotic usage and the development of type 2 diabetes, but have yet to confidently determine causality. Objective: To investigate the current precedent for excessive antibiotic usage in correlation to development of type 2 diabetes. Methods: This literature review was conducted using the framework of PRISMA to find causative associations between the extreme usage of various antibiotics and subsequent development of type 2 diabetes. Search terms initially used to identify synonyms included "antibiotic"AND "metabolic disease". In reducing the scope of the search, "antibiotic" AND "type 2 diabetes" were found in all relevant articles from the PubMed database. Results: Ten studies were analyzed after meeting the inclusion criteria. Each study indicated a statistically significant positive correlation between antibiotic use and a higher risk of developing type 2 diabetes. Conclusions: The common patterns of increased type 2 diabetes risk and excessive antibiotic use were: the type of antibiotic (quinolones, narrow-spectrum antibiotics), the number of antibiotic prescriptions (5 or more), and the duration of antibiotic use (medium- and long-term). Further studies will need to be conducted to determine causality, as the current leading interpretation suggests intensive antibiotic consumption decreases gut microbiome diversity, increasing the risk of obesity and type 2 diabetes.

      Gronewold, Laura; Hayden, Hannah Rae (The University of Arizona., 2024)
      Active shooter incidents in educational settings in America present a significant threat to the safety and well-being of students and faculty. The following is an overview of a research project focused on addressing active shooter safety measures at the University of Arizona in the aftermath of a tragic shooting incident in October 2022. Drawing on an engaged literature review and interviews with key stakeholders, this project aims to propose actionable recommendations to enhance campus safety. The process begins by contextualizing the prevalence of school shootings nationwide, highlighting the urgent need for proactive measures. Personal experience during the shooting incident served as a catalyst for the research, emphasizing the need for immediate action. Through collaboration with a faculty team and examination of successful initiatives at other schools and universities, the project proposes the implementation of Safety Preparedness Kits on the University of Arizona's main campus, which would be equipped with essential tools and informational resources in every classroom. Furthermore, it advocates for the development of QR code sites tailored to display classroom-specific safety information. Recommendations underscore the importance of not only proactive but also reactive measures to empower individuals and mitigate the impact of potential threats. The next steps I recommend to the University of Arizona present an opportunity for collaborative effort among students, faculty, and university leadership to prioritize safety and create a secure learning environment.