Alliance One Tobacco (Malawi)
Academic Research Repository at the Institute of Developing EconomiesNot a member yet
52305 research outputs found
Sort by
An Agent by Any Other Name: A Exploration of the Influence of Mental Models, Priming, and Perception on Users’ Views of Chatbot Character and Evaluation of Agent Effectiveness
No full textThis thesis endeavors to delineate the evolving interface between human cognition and artificial intelligence (AI), aiming to contribute to the corpus of knowledge within this rapidly advancing domain. It investigates the ability of psychological priming and the placebo effect to alter one’s mental models to enhance user interactions with AI technologies. Through a randomized control experiment, surveys initially assessed users’ attitudes towards AI and subsequently introduced supraliminal priming via descriptions of a purportedly unique AI mental wellness chatbot. The findings establish that priming users measurably impacts their experience with AI. A micro-diverse set of priming objectives (Avery vs. Charlie vs. Jamie vs. Bot characters) yield observable differences in response. Character priming was found to improve positive participant experiences relative to the neutral Wellness-Bot priming, to a similar degree as initial attitudes towards AI, which also enhanced the overall experience. This research underlines the importance of personalized character creation in the design and deployment of digital wellness interventions, as well as the essential role of fostering positive AI attitudes for successful integration of chatbots in user-centric applications. By investigating the overlap of cognitive science and AI research, this work provides foundational insights for fostering more intuitive and rewarding human-AI relationships, especially in enhancing mental wellness through digital interventions.AcknowledgmentsComputer Scienc
Identifying therapeutic vulnerabilities in RAS pathway–driven melanoma
No full textCutaneous melanoma is a malignancy that arises from melanocytes within the skin. Over 100,000 new diagnoses of melanoma are predicted within the United States in 2024, and incidence rates of melanoma have continued to increase, despite public health efforts to enhance melanoma prevention and management. While immune checkpoint blockade has revolutionized the standard of care and clinical outcome for advanced melanoma, many tumors are intrinsically resistant or often become resistant to immunotherapy over time, highlighting the urgent need to identify and develop novel therapeutic strategies to provide clinical benefit for patients with melanoma. The majority of melanomas are characterized by oncogenic driver mutations in genes within the RAS–MAPK signaling pathway, including BRAF, NRAS, and NF1, and may respond to targeted inhibition of this pathway. Indeed, combined inhibition of BRAF and MEK has been approved for the treatment of BRAFV600-mutant melanoma, although disease progression is common. Moreover, MEK inhibitors are not effective as single agents in NRAS-mutant melanoma and no other targeted therapies have been approved for these malignancies. To this end, this dissertation describes a genome-scale investigation of genetic pathways that sensitize RAS pathway–mutant melanoma to RAS pathway inhibition, leading to the development of two distinct approaches to induce melanoma cell death. Through the analysis of a genome-scale CRISPR–Cas9 knockout screen in NRAS-mutant melanoma cells, we identified the deubiquitinating enzyme USP7 as a target that elicits potent antitumor effects in vitro and in vivo when co-suppressed along with MEK inhibition. Integration of functional genetic approaches with epigenetic and transcriptomic analyses revealed that inhibition of USP7 in the context of RAS pathway inhibition influenced melanoma cell state in some settings and led to the upregulation of multiple transcriptional programs that contributed to melanoma cell death in NRAS-mutant as well as BRAF-mutant melanomas. This dissertation also describes the identification of the E3 ubiquitin ligase MARCH5 as a distinct vulnerability in NRAS-mutant melanoma that may be targeted in combination with MEK inhibition to promote melanoma cell death, likely by converging on a process of mitochondrial morphology and homeostasis. This dissertation concludes with a comprehensive overview of preclinical and clinical literature and categorizes combinatorial strategies that may be effective in RAS pathway–driven tumors of other tissues of origin. In summary, we have identified two distinct therapeutic vulnerabilities in RAS pathway–mutant melanoma and have provided a conceptual framework for future development of combination therapies in the treatment of melanoma and other RAS pathway–driven tumors.Medical Science
From Temperature Extremes to Ocean Circulation: Insights from the Transport of Tracers
No full textTracers are ubiquitous in both the atmosphere and the ocean. This dissertation develops a tracer-transport-based perspective to provide a new quantitative toolbox and novel insights into two key problems in atmospheric and oceanic sciences.
The first problem concerns atmospheric temperature extremes (Chapter 1). By treating temperature as a tracer governed by its transport equation, I propose a new framework to quantify how different physical processes control the probability distribution functions (PDFs) of temperature at any given location. Using this framework, we find that horizontal temperature advection is the dominant control on local temperature extremes, governing extreme events over the mid-to-high-latitude oceans and approximately 60 % of the global land area. We further use this framework to decompose the contributions of different eddy and mean components of advection to the global distribution of temperature standard deviation and skewness. In particular, we show that the eddy-mean interaction term, v′ · ∇ T̄, makes a significant contribution to skewness—a contribution that has been largely overlooked in previous studies. This term highlights the potential amplification of temperature skewness due to changes in the waveness of atmospheric circulation and polar amplification. Finally, we briefly discuss the broader implications of this framework: it can be extended to understand the spatial distribution of other key tracers in the atmosphere and ocean, such as pollutants.
The second problem addresses the relationship between ocean circulation and the spatial distribution of ocean water age (Chapters 2 and 3). Starting from the age-tracer transport equation, I derive a conservation law for age: for any control volume, the total age flux across its boundary equals the volume of the control domain (Chapter 2). This conservation law allows us to infer diapycnal mixing and overturning circulation directly from the spatial distribution of age. Using a set of idealized single-basin MITgcm simulations, we demonstrate that the global overturning rate across different density levels can be quantitatively estimated from the age difference between upwelling and downwelling regions. We further show that this framework can be applied to regional ocean circulation. For example, by analyzing the zonal-mean vertical profile of age at the equator, we estimate the deep overturning rate in the North Pacific, yielding results consistent with previous studies.
This age framework is also applied to investigate the physical controls on the ventilation of mid-depth North Pacific waters—the oldest waters in the modern ocean (Chapter 3). We find that diapycnal mixing is the dominant control on ventilation in this region, while isopycnal mixing and overturning circulation play secondary roles. Based on this finding, we propose a method to estimate basin-scale effective diapycnal diffusivity from the vertical profile of horizontally averaged age. Applying this method to observational radiocarbon data, we obtain a bottom-enhanced diffusivity profile with magnitudes on the order of 10⁻⁴ m² s⁻¹, consistent with previous estimates. However, our inferred diffusivities are somewhat larger than those from state-of-the-art bottom-up estimates, highlighting the uncertainty in basin-scale diffusivity and the need for further investigation. The three applications of the age framework—global overturning, deep North Pacific circulation, and mid-depth North Pacific ventilation—demonstrate the potential of the age framework to enhance our understanding and quantification of past and present ocean circulation.Engineering and Applied Sciences - Engineering Science
Zealots of Justice: Coercive Officials in Late Medieval Italy
No full textThis dissertation considers how, between about 1200 and 1350, cities in the upper half of the Italian peninsula experimented with a new kind of official whose primary task was coercion. There is a rich scholarship on institutions of public justice in the self-governing “communes” of thirteenth- and fourteenth-century Italy, but little of it has engaged with the coercive apparatus that served the courts and their users. It was easily the most sophisticated in Europe since the fall of the western Roman Empire. Italy’s coercive officials originated as messengers—comparable to process servers in the modern American legal system—in the years around 1200. But as communal courts sponsored new procedures that enabled creditors to swiftly recover debts, court messengers were tasked with coercive services for which creditors could pay—namely, the seizure of debtors’ property or, if necessary, the arrest and incarceration of debtors themselves.
The earliest coercive officials, although known by different terms from city to city, closely resembled one another by the end of the thirteenth century. I call this prototypical coercive official a “sergeant.” Sergeants had their analogues elsewhere in the medieval Mediterranean, even in northern Europe, but the surpassingly rich archives of a few Italian cities make it possible to study sergeants at a near-ethnographic level of detail. I use one such city in western Tuscany, Lucca, as my central case study. Drawing upon the tools of institutional history, prosopography, anthropology, and social network analysis, I seek to explain why coercive officialdom took an especially brutal turn in Lucca, and possibly other cities like it, in the decades after 1300.
Chapter 1 examines the legislative compendia, or “statutes,” that defined coercive officialdom in sixty-five Italian cities between about 1175 and 1360. These reveal how the office became increasingly brutal as it responded to the demands of creditors and communal administrations. Chapter 2 turns to the case of Lucca, concentrating on the 1330s and early 1340s. Here, we can see how the dictates of the office could be tempered by the social environment in which sergeants—called “heralds” (nuntii) in Lucca, as in other Tuscan cities—were embedded. However, the Lucchese evidence also reveals how the technology of coercive officialdom could directly serve the state. In Lucca, the exaction of taxes, fines, and other debts owed to the commune itself was not left to sergeants, but to officials called “retainers” (familiares). I use digital techniques to study the social networks that formed among sergeants, retainers, and the sureties they presented for office, showing that these two communities of coercive officialdom experienced the city around them in fundamentally divergent ways. Retainers, unlike sergeants, were largely excluded from Lucchese society and free to be shaped into “zealots of justice,” to use the memorable expression that one retainer applied to himself and his colleagues. In practice, “zeal” often meant brutality.
Chapter 3 delves into the world of retainers and their culture of brutality, which I argue was fostered by the institutions they served. Indeed, even as the Lucchese commune made a show of condemning the worst of this brutality, a regime of impunity prevailed. Chapter 4 shifts perspective and considers the resistance that brutality engendered, especially in Lucca’s countryside. Examining a decade of resistance, from 1334 to 1344, I show that the Lucchese commune authorized techniques of coercion that evoked the predatory tactics of marauding armies. Brutality of this kind often met with collective, organized resistance. While this dissertation charts the intensification of state-sponsored violence in the cities of late medieval Italy, it also shows that the trend toward brutality was far from inevitable or sustainable.Histor
Archaeological Insights into the Role of Religion in Funerary Rituals of Early Colonial Delaware: The Elkins Site Burial Ground Case Study
No full textThe Elkins site burial ground is located in New Castle, Delaware. The first record
of land ownership is from 1864, when Amos Nichols purchased the property. The Elkins
site comprises a house site and a small burial ground, consisting of five sets of skeletal
remains: one sub-adult male, two adult females, and two adult males. The ancient DNA evidence suggests that all five skeletons belonged to people of European descent. The Delaware Valley was a region of significant European migration during the seventeenth
and eighteenth centuries. There is a lack of research on the burial practices of early Americans within the Delaware Valley. By further examining the burials discovered at the Elkins site burial ground, it is possible to gain a deeper understanding of the role of
religion in these funerary rituals. This thesis focused solely on the remains located in the burial ground at The Elkins site burial ground. The burials were examined to compare with other cemeteries
dating to the same time frame, as well as nearby cemeteries. The primary focus was on the positions in which the individuals were buried, hand placement, burial shrouds, as
examined through the analysis of shroud pin usage, as well as the shape and lid style of
the coffins in which the individuals were interred. The Elkins burial ground yielded no grave goods other than pins, which were identified as shroud pins. By examining other nearby burial grounds, it was believed that this thesis would provide more information about the role of religion in funerary rites within eighteenth-century Delaware; however,
the available data was insufficient to draw a definitive conclusion.Extension Studie
Redesigning Opportunity: Creating Visual Tools for Low-Income Parents to Understand Socioeconomic Mobility
No full textThis thesis addresses declining socioeconomic mobility in America by developing a user-centered
visualization tool for low-income families. Despite research showing how geography impacts children’s outcomes, existing tools like the Opportunity Atlas primarily serve researchers rather than
families making critical decisions. Through a six-phase approach—need-finding interviews, thematic analysis, prototype development, co-design, enhancement, and testing—we created OpportunityViz, a web application translating mobility data into accessible guidance. Through interviews
with first-generation, low-income parent-child pairs, we identified ten themes shaping mobility
experiences and developed corresponding features. Usability testing showed features providing personalized recommendations, cultural considerations, and concrete action steps were highly effective.
Findings reveal effective mobility tools must offer personalization, cultural sensitivity, actionable
guidance, multiple pathways, and language accessibility. This research demonstrates how technology can bridge academic knowledge and practical application for vulnerable populations. By making mobility information accessible, OpportunityViz can empower disadvantaged families to make
informed decisions about their children’s futures, contributing to greater educational equity. A
demo of the tool is available at https://opportunity-ai-omega.vercel.app/Computer Scienc
COMPARATIVE MICROVASCULAR PROFILING IN SPEECH-RELATED BRAIN REGIONS USING ARTERIAL SPIN LABELING MAGNETIC RESONANCE IMAGING: A STUDY OF HEALTHY INDIVIDUALS AND MOYAMOYA PATIENTS
No full textCerebral perfusion, which represents the delivery of oxygen-rich blood to the brain, is vital for maintaining cognitive and neurovascular function. Critically, it progressively declines with age and in the presence of cerebrovascular disease. Arterial spin labeling (ASL) magnetic resonance imaging (MRI) is a noninvasive imaging technique that uses radiofrequency labeled arterial blood water as an endogenous tracer to measure cerebral blood flow (CBF; measure of perfusion) and arterial transit time (ATT; the time it takes blood to reach brain tissue). ASL is well-suited for studying microvascular changes across the lifespan and in clinical populations, due to its quantitative, safe, and repeatable nature.
This study leverages ASL to examine CBF in cortical regions involved in speech and auditory processing across two populations: (1) neurologically normal adults from the Human Connectome Project–Aging (HCP-A), and (2) patients with MMD, a rare but severe cerebrovascular condition, from the Vanderbilt Assessment of Multi-Model MRI Patients at Risk for Stroke with Intracranial Stenosis (VAMMPRIS) dataset.
In the healthy cohort, aging was associated with widespread declines in CBF and increases in ATT in left-lateralized language regions. Even in the absence of overt disease, individuals taking antihypertensive medications exhibited significantly reduced CBF and delayed ATT in multiple regions, highlighting the impact of systemic vascular burden on brain perfusion. Individuals with Moyamoya Disease (MMD) showed more pronounced reductions in CBF across all speech-related cortical regions, reflecting severe hypoperfusion. Sex-stratified analysis revealed that males with Moyamoya exhibited a steeper age-related decline in CBF than their female counterparts.
Detection of subtle and severe perfusion deficits with ASL MRI demonstrates its potential as a tool for monitoring cerebrovascular health across the spectrum of aging, hypertension, and disease. By characterizing how CBF is altered in specific brain regions, this study contributes to a better understanding of how vascular risk factors shape neurovascular aging and highlights the clinical value of ASL in identifying individuals at risk for cerebrovascular conditions.Biomedical Engineering A
Endothelial Gap Junctions in the Control of Neurovascular Coupling
No full textThe brain depends on a highly regulated blood supply to power the energetically expensive computations that underlie cognition. To ensure that energetic resources are efficiently allocated, the brain dynamically redistributes blood flow to active regions via a process known as neurovascular coupling. This close matching of neural activity and hemodynamics is thought to be essential for neuronal homeostasis and forms the basis of non-invasive functional brain imaging in humans. That said, the molecular mechanisms underlying neurovascular coupling remain poorly understood.
To generate robust changes in local perfusion, neurovascular coupling involves the rapid, coordinated dilation of large stretches of the brain’s arterial network. The goal of this dissertation is to understand how this finely tuned, moment-to-moment matching of neuronal and vascular systems is achieved. Specifically, we demonstrate that endothelial gap junction-mediated signaling serves as an intermediary linking neuronal activation to long-range mobilization of the arterial network.
Leveraging a novel, endothelial-specific adeno-associated virus variant, we develop a non-invasive methodology to assay gap junction coupling in vivo. Using this technique, we find that endothelial cells throughout the central nervous system are functionally interconnected by gap junctions. Furthermore, we find that both the strength of this coupling and the specific connexin isoforms used by endothelial cells to form gap junctions vary along the arterio-venous axis. Based on these results, we generate a conditional loss-of-function mouse model to acutely abolish arterial endothelial cell gap junction coupling in the cerebrovasculature. Finally, using a combination of visual and optogenetic stimuli presented to awake mice, we show that endothelial gap junction coupling is essential for rapid, long-range propagation of vasodilation during neurovascular coupling.Medical Science
More Than Meets the Eye: Is It Feasible To Uncover Intent From Saccade Sequences?
No full textIn 1967, Alfred Yarbus demonstrated that human eye movements are influenced by cognitive tasks, suggesting they can reveal an observer’s intent. While subsequent research produced mixed results regarding this claim, advancements in machine learning provide new ways to explore the predictability of intent from eye movement data. This thesis revisits this topic, focusing on whether saccade sequences can predict intent beyond random chance and what patterns models prioritize in these predictions. Our findings confirm that intent can be reliably predicted above chance when models are trained on targeted data, such as distinguishing between tasks where a target is present (TP) or absent (TA). Performance improvements were noted when the models specialized in either condition, although cross-scope generalization remained challenging. Temporal dependencies emerged as crucial, demonstrated by a significant accuracy drop when saccade sequences were reversed, highlighting the role of sequence order. Additionally, spatial occlusion experiments showed that specific image regions contribute differently to prediction confidence, with context-based models leveraging broader scenes and detail-based models focusing on identifiable objects. These results support Yarbus' hypothesis of purpose-driven eye movements, underscoring the intricate interplay of spatial and temporal elements in intent prediction.Computer Scienc
Effects of Pd ensemble size in dilute and single atom alloy PdAu catalysts for one-pot selective hydrogenation and reductive amination
No full textIn the one-pot reaction between nitroarenes, aldehydes, and hydrogen, the desired outcome is the selective hydrogenation of nitroarenes to form aminoarenes that condense with aldehydes to form pharmaceutically relevant imines and N-alkylamines. One approach to facilitate the selective hydrogenation of nitroarenes over aldehydes involves using bimetallic catalysts with near equimolar ratios. However, structural characterization of metallic ensembles on the nanoparticle surface is challenging at such high alloying ratios, which hinders the elucidation of clear structure–property relationships. Here, we prepared a well-controlled series of dilute Pd-in-Au alloy catalysts with a fixed nanoparticle size as a model system to investigate the effects of surface Pd ensemble size, from single atoms to dimers and trimers, in the one-pot hydrogenation reaction between nitrobenzene and benzaldehyde as our probe reaction. The highest (near unity) selectivity to condensation products was achieved using the catalyst with the lowest Pd content prepared (Pd2Au98/SiO2), which predominantly exposed surface Pd single atoms as verified by surface-sensitive spectroscopy. Theoretical calculations reveal that Pd single atoms were inactive for benzaldehyde adsorption and thus enabled selective nitrobenzene hydrogenation. On the contrary, the adsorption of benzaldehyde became stronger than nitrobenzene for Pd trimers and larger ensembles, explaining the enhanced competitive adsorption from benzaldehyde with increasing Pd content. Our results demonstrate that the commonly used (near equimolar) alloying ratio is rather arbitrary and may not necessarily produce the highest selectivity to condensation products. Instead, we illustrate how nanoscale Pd ensemble size control tunes competitive kinetics to steer selectivity towards forming the desired condensation products.Chemistry and Chemical BiologyMaterials Science and Mechanical EngineeringAccepted Manuscrip