33760 research outputs found
Sort by
DEVELOPING TOXICOLOGICAL TESTING AND FECUNDITY ASSESSMENT METHODS USING ESTUARINE FISH TO MONITOR LEGACY CONTAMINATION
No full textEstuarine sediments bear the legacy of urban and industrial pollution, where benthic organisms and their predators are exposed to persistent organic pollutants (POPs) such as Polycyclic Aromatic Hydrocarbons (PAHs), Polychlorinated Biphenyls (PCBs), and Dioxins. We are using the mummichog (Fundulus heteroclitus) to study fish health in historically contaminated urban tidal rivers. This thesis contains four chapters: an introductory literature review, studies developing embryo-larval sediment contact assays for mummichog and sheepshead minnow (Cyprinodon variegatus), methods for assessing fecundity in female mummichog sourced from a historically contaminated urban river, and general conclusions. The introduction provides background to sediment contamination and discusses the rationale for the use of mummichog as a species of interest for toxicological assessment. It also proposes the use of sheepshead minnow as a surrogate to assess sediment toxicity and facilitate comparison with existing literature. The second chapter describes in detail a lab methodology for an embryo-larval whole sediment toxicity assay with mummichog and sheepshead minnow and discusses the results of tests with each species to spiked sediment exposures. The proposed approach modifies an established water column toxicity test, exposing fish embryos to three POP contaminants in spiked sediment dilution series. The contaminants used in the sediment spiking belong to more general classes of POPs: PAHs, PCBs, and Dioxins, and are commonly found in degraded aquatic habitats. The specific contaminants used in the exposures are Creosote, Aroclor 1254, and 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD), and their impact is assessed using mortality, health condition, growth, and time to hatch. The results indicate that both species are amenable to toxicity testing with whole sediment exposure, with different species sensitivity as reflected in variable endpoint responses. The third chapter explores fecundity methods as a proxy of environmental effects on mummichog to expand biomonitoring beyond histopathology. This effort applies gonadosomatic index (GSI) and oocyte enumeration methods, throughout the reproductive season to mummichog collected from four spatially distinct populations in an urbanized tidal river. Overall, this work aims to further establish mummichog as a site-relevant species for toxicity testing, both in the lab and biomonitoring
THE ROLE OF TASK DEMANDS, PROFICIENCY, AND ORTHOGRAPHY IN BILINGUAL LANGUAGE SWITCHING
No full textThe current study explores language switching in bilinguals and investigates how task demands, proficiency, and orthography contribute to switching costs and patterns between languages. Korean-English bilinguals participated in this study and completed a bilingual naming task in either Hangeul/English or Korean romanization/English, a language-inclusive lexical decision task (also in Hangeul/English or Korean romanization/English), a language background questionnaire, and two proficiency measures: an English cloze test and the English Elicited Imitation Test (EIT).
The bilingual naming task was designed to measure language switching in production, while the lexical decision task assessed switching in comprehension. The task was visually the same, however, participants were asked to name words in one task and recognize words in the other. Participants were assigned to either the Hangeul/English or the Korean romanization/English condition and completed both tasks within their assigned condition. All seventy-five participants included in this study achieved at least 70% accuracy on both tasks. At the time of the study, all were living in an English-speaking country, with Korean as their first language (L1) and English as their second language (L2).
In the naming task under the Hangeul/English condition, switching costs were greater when switching from L2 English to L1 Korean than in the reverse direction. However, in the version of the task with Korean romanization (instead of Hangeul), the pattern was reversed: switching from L1 Korean to L2 English incurred greater costs. In the lexical decision task, switching costs were symmetrical for the Hangeul group but asymmetrical for the romanization group, with greater costs observed when switching from L1 Korean to L2 English. Proficiency, as measured by the EIT, played a significant role in reaction times across most tasks.
These findings are discussed in relation to theoretical models of bilingual language control, with particular attention to how different task demands, orthographic representations, and L2 proficiency modulate switching costs
“To be Pure or Prosperous”: The Archaeology of Transient Laborers and Saloons along the Minnesota and North Dakota Border (1870-1940)
No full textThis dissertation expands how archaeologists conceptualize marginalized groups like transient laborers consisting of loosely grouped historical characters like the hoboes, tramps, migrant workers, and others as they moved through highly concentrated and contested spaces. To do this I combine the archival record with the archaeological record of the Saloon Row in Moorhead, MN. I argue that saloons acted as “nodes of visibility” in which the embodied marginalization of these transient laborers comes to the forefront in a moral economy. Moreover, I examine this process of capitalism in terms of production, via labor on industrial bonanza farms and consumption, via saloons. This dissertation seeks to examine primarily three research questions: 1) How does the material culture of saloons reflect the presence of transient laborers? 2) What does this tell us about the intimate connection between production and consumption in a moral economy? And 3) Is there any connection between transient laborers and contemporary unhoused people
UNDERREPRESENTATION OF BLACK STUDENTS IN HIGHER LEVEL SECONDARY MATHEMATICS COURSES: AN ANALYSIS OF THE IMPACT OF REMOVING QUALITATIVELY INFERIOR MATHEMATICS COURSES ON STUDENT ENROLLMENT, COMPLETION, AND ACHIEVEMENT
No full textThis dissertation investigates the underrepresentation of Black students in higher-level secondary mathematics courses within Calvert County Public Schools (CCPS). The study addresses the persistent opportunity gap between Black and White secondary students, a disparity historically perpetuated by district practices that tracked Black students into lower-level mathematics courses. In response, a district initiative—guided by principles from the National Council of Teachers of Mathematics (NCTM) and supported by a small group of committed district staff—sought to increase Black student enrollment in advanced mathematics courses. The initiative aimed to enhance mathematical achievement and literacy, both of which are critical for future academic and career success.
Employing a quantitative descriptive, ex post facto design, the study analyzed administrative data from CCPS on course enrollment, completion, and achievement (primarily SAT Mathematics scores) for freshman cohorts from 2006, 2011, 2016, and 2021. Additionally, an online survey was administered to all high school mathematics teachers to gather their perspectives on the district’s decision to eliminate certain lower-level high school mathematics courses—Prealgebra, Intermediate Algebra, Algebra III, and Business Math—beginning 2014. The study examined both the impact of these course eliminations and the evolution of teacher perceptions over time.Key findings indicate that the removal of these courses was associated with increased access to Algebra II and subsequent University System of Maryland (USM)-approved post-Algebra II courses, particularly for Black students, increasing from 51% of Black students earning credit for Algebra II to 79% and increasing from 22% of Black students earning credit for a USM-approved post-Algebra II course to 69% over the two decades. Moreover, Black students demonstrated a significant increase in SAT Math scores relative to their Reading scores and to the Math scores of the overall student population, with cumulative average annual increase rates of 4.2%, 9.6%, and 13.9% respectively at the 400+, 500+, and 600+ levels, compared to 1.7%, 3.6%, and 3.3% for the full population, suggesting improved achievement outcomes linked to increased access to rigorous coursework. Teacher perceptions, initially skeptical of the course eliminations, shifted positively after reviewing data showing increased enrollment and achievement among Black students.
These findings have important implications for educational policy and leadership. They suggest that systemic course restructuring can be an effective strategy for addressing racial disparities in mathematics education. The study contributes to the broader understanding of how intentional policy changes can promote equity and expand opportunities for historically underserved student populations in CCPS, across Maryland, and potentially in other districts nationwide
INVESTIGATION OF SCRAMJET PERFORMANCE DURING COMBUSTION MODE TRANSITION
No full textCombustion mode transition is a key challenge in dual-mode ramjet engines (DMRJ) due to its direct effect on the engine’s performance and operation. DMRJs are designed to operate both in a ramjet mode at lower supersonic speeds and in a scramjet mode at higher speeds. The investigation of the engine’s performance during the combustion mode transition is desirable to ensure continuous engine operability over the transitional Mach numbers. Depending on the flight speeds, the combustion mode can switch from subsonic combustion mode, commonly known as thermally choked mode, to supersonic combustion mode and vice versa. This study investigates and quantifies the effects of these transitions on combustor stability and propulsion performance metrics.The mode transition is expected to occur near the start of the hypersonic flight regime. Combustion mode transition experiments were conducted using a laboratory-scale, direct-connect test facility that employed a vitiated air heater to match the enthalpy of Mach 5 flight conditions at an altitude of 19 km. A convergent-divergent supersonic nozzle was used to establish a Mach 2 isolator entrance flow, which defined the upstream boundary conditions. The combustor model featured one fuel injector upstream of the cavity flame-holder and three more fuel injectors downstream. Previous studies have shown that natural mode transition occurs at or near the critical equivalence ratio in a series of sporadic transitions back and forth over a relatively substantial length of time spanning around 400~600 msec. This study investigates and quantifies the effects of these transitions on combustor stability and propulsion performance metrics.
High-frequency dynamic pressure measurements and axial distribution of wall pressure were measured at various equivalence ratios near the mode transition conditions. The results showed that combustion instabilities are observed during the natural mode transition process at or near the critical equivalence ratio of 0.24. The dominant oscillations were observed at 957 ± 10 Hz with a peak amplitude of 0.2 psi. Also, a cycle analysis was performed to investigate the effect of mode transition on instantaneous thrust. Using previously obtained velocity data and wall pressure measurements, local thrust function was computed, and the cycle analysis was extended into an assumed pressure-matched exit nozzle. The cycle analysis results revealed a substantial effect of combustion mode transition on expected thrust level, changing the amplitude by more than 13%. Finally, a mode transition control strategy is discussed that can not only avoid the combustion instability but also reduce the sudden disruption caused by thrust change associated with the mode transition
CONTEXT-DRIVEN EXPECTATIONS IN REAL-TIME SENTENCE PROCESSING: INVESTIGATIONS OF ADULTS, CHILDREN, AND LARGE LANGUAGE MODELS
No full textA central feature of human language processing is the ability to rapidly build and update complex representations based on prior context. While humans actively use various linguistic cues to guide expectations, they sometimes fail to show immediate sensitivity to contextual information. A striking example is the role-reversal phenomenon, where comprehenders appear to expect verbs that fit the opposite argument roles indicated by the preceding context. These errors have often been attributed to failures in context-based representation building processes.This dissertation investigates how context-driven expectations unfold in real-time, focusing on the mechanisms that support anticipatory processing and where vulnerabilities arise. Using the role-reversal phenomenon as a test case, I unpack the generation and integration processes into distinct stages, in order to determine which parts are robust and which are more prone to errors. Through a combination of behavioral, electrophysiological, developmental, and computational methods, I examine divergences in role-sensitivity between comprehension and production measures, between adults and children, and between humans and computational models of language processing. The findings overall demonstrate that context-driven representations are generally robust and that errors tend to arise in later stages—particularly when new input must be integrated into prior representations under conditions of weak expectations. Moreover, the variability across different measures, sentence types, and populations—including children and large language models—can be traced to differences in the strength and timing of these underlying processes.
Together, the results indicate that predictive failures do not reflect a general weakness in the representation-building system but rather stem from specific points of vulnerability. These insights help to refine our understanding of how context-driven expectations operate in real-time and contribute to broader theories of the cognitive mechanisms that support human language processing
ARTIFICIAL INTELLIGENCE FOR ACCELERATING AND UNDERSTANDING MOLECULAR SIMULATIONS
No full textComputational techniques such as molecular dynamics (MD) simulations offer detailed spatiotemporal insights into biomolecular systems, playing a critical role in uncovering mechanisms and informing therapeutic design. However, a major limitation of MD is its high computational cost, which makes it challenging to study many biophysically relevant processes.In this dissertation, I address this challenge by integrating artificial intelligence (AI) with statistical physics to develop enhanced sampling methods that significantly accelerate MD simulations. These physics-driven, representation learning approaches enable efficient implementation of molecular dynamics in systems that would otherwise be computationally intractable. Moving beyond traditional model systems, I demonstrate the practical utility of these methods in drug discovery, particularly in characterizing the interactions between small molecules and RNA, an emerging class of therapeutic targets.
Furthermore, such AI-driven approaches typically operate in data-sparse regimes and it is essential to establish the robustness of the trained models before deploying them. For this purpose, I design an algorithm to validate general-purpose AI models, particularly in the context of MD.
Overall, this dissertation demonstrates how the principled integration of AI with physics-based computational methods enables more efficient and insightful molecular simulations
EXPERIENCING CASTE: SOCIAL DISTANCE IS DISTINCT FROM ECONOMIC DISTANCE
No full textCaste has been a primary identity factor in the Indian subcontinent for thousands of years. It originated from a religious framework, coded by early law books, but as society shifted with different rulers, political systems, and policies, the influence of caste in maintaining the distances between groups has changed. As a society changes, the power structures between different groups within the society inevitably shift, impacting the experience, outcomes and behavior of dominant and oppressed groups in differing ways. To study the impact of societal change on caste groups, I argue that caste inequalities should be observed and operationalized in social terms separately from economic terms, though both social and economic lenses are important. Moreover, I also argue that it is important to examine the relationship between these two axes of stratification by locating them in local rather than national contexts.Social caste distance is related to ritual beliefs and associated practices that maintain a social distance between different caste groups in a locality. Economic caste distance is related to the material wealth inequalities that exist between caste groups living in the same locality. This study serves to highlight that social and economic caste distance should not be considered one and the same, though they may interact. Using the India Human Development Survey Wave 2 (2011-12), I structure my study at the local area level, generating district-level markers for both the social and economic distance between different caste groups to operationalize the caste context in which each family lives its life. I test how the social and economic distance may impact dalits (also called Scheduled Castes) and dominant castes differently by looking at three outcomes: Local conflict, women’s labor force participation, and children’s learning outcomes.
I find that social distance predicts conflict slightly better than economic distance, but places with both social distance and economic distance are even more likely to have conflict. For women’s labor force participation, I find that there are significant differences between dalit women and dominant caste women’s behavior. All women are less likely to work in areas with high social distance, but dalit women are likely to work more in areas with high economic inequality between caste groups despite social distance present in the area. Dominant caste women are not as likely to work, compared to all other caste groups, and economic inequality between castes benefits dominant caste households and decreases a dominant caste women’s likelihood of working. I also find that the learning outcomes of children significantly differ among different caste groups. For both oppressed caste and dominant caste children, the variables that are significant when it comes to learning outcomes seem to be the socio-economic factors apart from the caste distance like household education, assets, and in the case of reading, the urban/rural status. There is no conclusive finding for all children taken together when it comes to the effect of economic or social distance between castes, BUT dalit children analyzed separately face a negative impact on both reading and math outcomes in areas where there is high economic distance between castes
GAMMA RAY BURST FOLLOW UP USING GROUND BASED METHODS: INSTRUMENTATION, OBSERVATION AND ANALYSIS
No full textSince the 1973 announcement of the initial Vela detection of Gamma-Ray Bursts (GRBs) in 1967, these powerful events have been of significant interest. The study of GRBs has advanced our knowledge of plasma physics, particle physics, cosmology, among many other fields. However, even with the advancement in understanding gained through various detections, both multi-wavelength and multi-messenger, there is still much to be learned by observing these transients in as many different ways as possible. This dissertation will explore the development of instrumentation (RIMAS and PRIME), operational software, data reduction software, and observations aimed at furthering the study of GRBs.
Rapid infrared IMAger Spectrometer (RIMAS) is a near-infrared (NIR) imager spectrometer operating in wavelengths from 0.97 to 2.37 microns, and will be installed on the 4.3 m Lowell Discovery Telescope (LDT) in Happy Jack, Arizona. Among other features, RIMAS has four broad band filters (Y, J, H, K), and three spectral modes (spectral resolving powers, R~30, 300 and 5000). Completing this instrument required the development, integration and testing of multiple detector readout systems, cryogenic systems, spectral and imaging modes, communication systems, and cooling systems. Furthermore, software needed to be developed to reduce RIMAS data so that it will be ready to contribute to scientific discoveries shortly after commissioning. Due to the development of this software, observers will quickly obtain high signal-to-noise exposures, astronomically and photometrically calibrated stacked images, and one-dimensional spectra that are flux and wavelength calibrated.
PRime-focus Infrared Microlensing Experiments (PRIME) is a 1.8 m telescope with 1.56 square degree field of view (FOV) (0.5”/pixel) operating in the NIR (0.82-1.78 microns) located in Sutherland, South Africa at the South African Astronomical Observatory (SAAO). The primary science objective for PRIME is to perform a time-domain survey of the Galactic Bulge for the purpose of observing microlensing events caused by transiting exoplanets. In doing so, it will support the Nancy Grace Roman Space Telescope (RST) mission by taking baseline measurements of this field before RST launch, while testing the performance of RST package H4RG-10 detectors. After RST launch, PRIME will perform simultaneous observation with RST of the Galactic Bulge to perform parallax measurements. PRIME is also continuously available for Target of Opportunity (ToO) observations, making it a powerful tool for observing GRB and gravitational wave (GW) counterparts. Since commissioning in October 2022, PRIME has been observing a time domain survey of the Galactic Bulge searching for microlensing events, observing high-energy transients (GRBs, X-ray bursts, GWs, and supernovae), and performing an all-sky survey in J-band. During the course of this thesis work, the development of PRIME's camera greatly mirrors that of RIMAS, with lessons learned shared between them. Most notable in this development is the integration and testing of PRIME's detectors and detector readout system. This is the first instance of these detectors and electronics being used on-sky, and as such required the development of custom software and thorough testing.
Beyond these instrumentation efforts, photometric and spectroscopic observations were performed in the optical for various GRB afterglows, GRB host galaxies, and GW counterpart candidates using LDT's optical imager, Large Monolithic Imager (LMI), and optical spectrograph, DeVeny. After the commissioning of PRIME, over 40 high-energy transients (GRBs, X-ray Bursts, LIGO/Virgo/KAGRA (LVK) probability fields) were observed with the new telescope. These events were observed as soon as possible after initial detection, and observed over multiple epochs when assessing the evolution of detected counterparts. Further observations were also obtained when sources were not well localized, and further observations could lead to the detection of variable sources. These have resulted in the PRIME detection of the brightest GRB of all time, GRB 221009A, the night after PRIME's first light, and helped determine that this extreme GRB was powered by a structured jet. PRIME observations of GRB 230307A helped determine that the progenitor system was a lanthanide rich kilonova despite its long duration. Observations and software developed for LDT during this thesis work searched for and categorized potential optical GW counterparts, analyzed the environments surrounding GRBs, and searched for a supernova associated with GRB 221009A
Heritage Speakers and Learners of French at the Tertiary Level: Curricular Implications
No full textThis dissertation utilizes qualitative methodology to investigate the experiences and needs of heritage speakers of French in higher education through a locally focused study at the University of Maryland, with the broader aim of contributing to a national understanding of this population of speakers. While heritage language education has gained momentum in the last couple of years, French heritage speakers remain underrepresented in research and curriculum development. Typically, these students demonstrate strong oral and aural competencies but have more limited literacy skills in the heritage language, particularly in writing. This study places students at the center by identifying their particular cultural and linguistic needs in mixed-learner classrooms, where students who acquired the language at home and those who learned it formally in school study together, often with significantly differing language proficiencies. It explores how heritage learners’ linguistic, cultural, and professional goals can inform curriculum design and guide the development of needs-based instruction. A needs assessment survey was conducted to examine students’ priorities and perceived gaps in their learning experience. Although cultural identity was expected to be a central motivator, the findings revealed that students were primarily driven by the desire to strengthen their French for professional and social purposes. Respondents also expressed strong interest in explicit instruction in grammar and writing. The survey findings inform ways to better support these French heritage students’ learning process, leading to suggestions for inclusive curricular adjustments. Drawing on task-based and content-based, this dissertation proposes a framework adaptable to other postsecondary institutions