Open Research Oklahoma (Oklahoma State Univ.)
Not a member yet
42035 research outputs found
Sort by
Peer disagreement, permissivism, and deference
No full textHow should an agent respond to peer disagreement? Two options have been defended in the literature: either remaining steadfast, where an agent maintains her credence, or conciliating, where an agent moves her credence closer to her peer’s. In this thesis, I defend a third option: deference. In chapter II, I argue that, in some cases, it is permissible for an agent to respond to peer disagreement about whether p by adopting the peer’s credence with respect to p. My argument draws on Kelly’s insight regarding permissivism, the claim that two agents are rationally permitted to have different responses to the same body of evidence. Kelly argues that accepting permissivism allows one to reject conciliationism. Kelly is right about that, but rejecting conciliationism does not lead directly to steadfastness. Permissivism also makes room for deference. Agents can reasonably differ on the extent to which they prioritize their own epistemic agency over others’ epistemic agency. Agents who value the epistemic agency of others more strongly are rationally permitted to defer to their peers in some cases of peer disagreement. In chapter III, I consider two strong objections to my account. The first is that agents should bracket their first-order evidence upon encountering the higher-order evidence that one’s peer disagrees. I respond by relying on an argument from Chen and Worsnip that peer disagreement has a role both as higher-order evidence and testimonial evidence. Building on this, I argue that it can be rational for an agent to place higher value in the testimonial evidence than she places in the higher-order evidence. Thus, it can be rational for an agent to defer as a response to peer disagreement. The second objection is that an agent should strike a balance between being a conformist and a nonconformist. I respond that an agent being more conformist can help her to achieve epistemic goods and that we should consider that some agents have the starting position of being more conformist
Novel interaction partners of PtsN, a regulator of type III secretion system expression in Pseudomonas aeruginosa
No full textThe opportunistic pathogen Pseudomonas aeruginosa (PA) causes a variety of infections and produces a large set of virulence factors contributing to its pathogenicity, notably the type 3 secretion system (T3SS), which aids in establishing and maintaining infections. The T3SS is used by bacteria to assemble a molecular apparatus that acts similarly to a hypodermic needle to inject effector proteins into the cytoplasm of host cells. Transcriptomic data revealed that the protein PtsN impacts T3SS genes. In strains with PstN absent, we observe substantial upregulation in T3SS gene expression. We hypothesized that PtsN may be interacting with the master activator of T3SS expression, ExsA, in a way that limits its ability to activate T3SS genes. Co-immunoprecipitation, western blotting, and mass spectrometry revealed no evidence of a PtsN-ExsA interaction but identified novel binding partners, including MucA, anthranilate synthesis enzymes, and the T3SS components PscU and PscR. Deletion of ptsN increases T3SS expression, but overexpression seems to have minimal effect, suggesting possible saturation of its regulatory impact under physiological conditions. These findings suggest that PtsN modulates the T3SS through a mechanism independent of a direct ExsA interaction and highlight new connections between PtsN and anthranilate metabolism
Victorian walking skirt
No full textThe Walking Suit was an outfit worn by women in the 1890s-early 1900s. I have designed and constructed a Victorian walking skirt for the theatre which included designing, patterning, constructing, fitting, and modeling the items myself
Comparison of Cover, Copy, Compare and Explicit Timing interventions to increase digits correct per minute scores
No full textAn adapted alternating treatments design was used to compare the effects of class-wide applications of Cover, Copy, Compare (CCC) and Explicit Timing (ET) procedures designed to enhance addition and subtraction fact fluency in an intact second and third-grade classroom, respectively. This dissertation also examined whether initial fluency level scores would differentiate intervention effectiveness between CCC and ET procedures. Results showed that ET caused greater increases in addition and subtraction fact fluency than a control condition. CCC did not result in higher learning rates than a control condition. The interaction between intervention type and initial fluency levels suggested mixed results. For both grades, students with fluency scores in the instructional range (11-20 DCPM), ET intervention resulted in higher DCPM scores than both CCC and the control; however, for students in the frustration range (≤10 DCPM), it was found that CCC did not outperform ET and the control conditions. The discussion focuses on using skill-by-treatment interactions to inform intervention selection, future directions for research, and addresses study limitations
Proactive path planning for human-centered robot navigation
No full textMobile robots that assist with physical therapy must be capable of tracking human subjects accurately and safely in dynamic, unstructured environments. A key challenge in this context is maintaining a desired distance and orientation relative to a moving person whose motion may be unpredictable. Achieving reliable and natural motion tracking is essential for enabling robots to support in gait-related clinical tasks without disrupting or constraining the patient’s movement. To address this, we propose a hybrid local planning framework that combines the Dynamic Window Approach (DWA) with a Genetic Algorithm (GA) to enhance trajectory selection for omnidirectional robots during human-following tasks. By replacing the standard trajectory sampling and scoring mechanism of DWA with a population-based search, the planner can better adapt to both curved and linear motion paths while maintaining relative pose constraints. The robot's motions is simulated using the Robot Operating System (ROS2), reflecting human movement patterns. The performance is evaluated in terms of positional accuracy, angle alignment, and path efficiency. The results show that our hybrid planner reduces tracking error and generates smoother, more human-aligned behavior than standard DWA, making it a promising approach for safe and responsive robot-assisted human gait analysis
Descending spiral: Case studies in political violence in early U.S. history
No full textThis research focuses on four distinct episodes of political violence in U.S. history that occurred during a thirty-year span from 1799 to 1829. The four events in question are Fries’ Rebellion of 1799, the German Coast Slave Revolt of 1811, the Baltimore Riots of 1812, and the Cincinnati Riots of 1829. The historiography on political violence in early U.S. history has, so far, been very siloed. Rather than considering different manifestations of political violence together, scholars have tended to isolate events like these from each other and have instead organized their analysis around specific time periods, methods of violence, or shared ideological motivations. The objective of this project is to utilize comparative analysis to bring these four different episodes of political violence into conversation with one another, examining their causes, the reactions by state authorities, and the depictions of them in the press. Some of the most useful sources for this research have been the personal papers of state and local officials, documents from legal investigations, newspapers, and travel narratives that provide insights into the communities in question. Through comparative analysis, this thesis argues that familiar motivations, such as economic frustrations and ethnic prejudices, could help generate acts of political violence in the early U.S. Additionally, government officials displayed a far greater willingness to forcefully confront those engaging in political violence when they directly challenged state authority. When the targets of the violence were private citizens and especially unpopular minority groups, authorities were more reluctant to get involved. The press also showed itself to be inconsistent in how it responded to different acts of political violence, with newspapers allowing their coverage to be influenced by numerous arbitrary factors, such as political affiliations, race, and international affairs. Ultimately, historical research on political violence is vital because it is a phenomenon that we still contend with today. Political issues change over time, but questions surrounding how political violence comes about, how authorities react, and how the press portrays such events are all relevant to the present
COVID-19, turfgrass preferences, and climate change perceptions: Exploring the role of demographics and weather variability
No full textChapter one investigates how the Covid-19 pandemic affected consumer preferences for turfgrass attributes, using survey data from January 2019 (pre-pandemic) and April 2021 (post-pandemic). A mixed logit model captures individual differences in preferences, and willingness-to-pay (WTP) estimates are compared across time. The model is re-estimated for risk-seeking and risk-averse consumers to determine the role of climate risk attitudes. A random-effects panel model further analyzes how demographics and risk attitudes influence WTP for improved turfgrass traits. Results show that WTP increased overall during the pandemic, particularly among risk-averse consumers, suggesting growing demand for drought-tolerant and stress-resistant turfgrasses in the face of climate change and health crises.
Chapter two examines how demographic characteristics and weather variability influence climate change perceptions. Using two surveys and a heteroskedastic ordered probit model, the study finds that political affiliation, age, education, race, and gender significantly shape attitudes, with Republicans, older adults, and less-educated individuals showing lower acceptance. Going beyond general attitudes, the chapter assesses concern for specific climate risks (e.g., drought, extreme heat) using linear, quadratic, lagged, and coefficient of variation models. Results show that temperature and precipitation affect concern in non-linear, seasonally distinct ways, particularly in summer and winter. National data from the Yale Program on Climate Change Communication (YPCCC) confirms these patterns, emphasizing consistent demographic trends and the importance of question framing. Even skeptical groups show modest concern when climate change is not directly referenced, showing the roles of weather patterns and demographic heterogeneity in shaping public perception
Optimization and tradeoffs in political redistricting using integer programming
No full textPolitical redistricting involves balancing many criteria such as population deviation, compactness, contiguity, and so on. These criteria often conflict, meaning that achieving the best performance on one criterion may require sacrificing some performance on another. Understanding these tradeoffs is key to drawing good redistricting plans. This dissertation proposes integer programming techniques to explore these limits and tradeoffs.
First, we consider the widespread belief among districting experts that drawing k equally populous districts requires at least k − 1 county splits. We demonstrate this belief to be false, finding counterexamples to claims from expert testimony that have made it all the way to Supreme Court oral arguments in Allen v. Milligan (2023).
Second, we examine the tradeoffs between criteria, focusing specifically on population balance and compactness. Previous researchers have used heuristic methods to approximate the associated Pareto frontier, but we propose to use exact methods. Our approach is based on the epsilon-constraint method and solves mixed-integer second-order cone programs (MISOCPs), at various population deviations, using a branch-and-cut algorithm. With these methods, we evaluate maps from the Supreme Court case Tennant v. Jefferson County (2012).
Third, we consider the US Census Bureau’s latest method for protecting respondent privacy and its impacts on redistricting. This latest Disclosure Avoidance System (DAS), known as the TopDown Algorithm (TDA), systematically adds random noise to the population counts, which has led to concerns that districts that appear population-balanced in the reported (noisy) data may not actually be balanced according to the underlying (pre-noised) counts. We propose mixed-integer programming (MIP) techniques that can be used to analyze the worst-case effects of such noising procedures. We find the effects to be minor when using counties or tracts to build districts but that huge discrepancies can arise when using census blocks. The effects are mitigated, but not eliminated, when districts must satisfy various districting criteria (e.g., compactness)
Examining the climatic effects of cropland expansion on natural grassland
No full textThis dissertation explores the climatic impacts of land use and land cover change (LULCC) through the lens of agricultural expansion into natural grasslands. Focusing on both biophysical and biochemical feedbacks, the research integrates multisource remote sensing data, eddy covariance flux tower observations, and the Transient Climate Response to Cumulative Carbon Emissions (TCRE) framework to assess land-atmosphere interactions across multiple spatial and temporal scales. In the northeastern Tibetan Plateau, green fodder plantations were found to reduce land surface temperature (LST) by an average of –0.54°C, with a peak cooling effect of –1.20°C in August, mainly driven by enhanced non-radiative process. In the U.S. Great Plains, four grass-to-crop conversion types showed significant cooling during the growing season (up to –1.98°C), with evapotranspiration alteration being the dominant mechanism. Spatial and seasonal variability highlighted the need to account for crop type and phenology in climate modeling. Furthermore, a phenology-based comparison of winter wheat and grassland using flux tower observations and MODIS LST data revealed stronger biophysical temperature effects (–1.10°C to +4.17°C) compared to negligible biochemical effects (<0.0001°C) due to limited regional carbon flux influence under full mixing assumptions. Collectively, the findings demonstrate that biophysical feedbacks dominate regional temperature responses to LULCC, while biochemical effects remain marginal at local scales. The study underscores the importance of crop-specific and sub-seasonal analysis, and offers a scalable framework to improve the integration of agricultural land management into climate impact assessments