114333 research outputs found
Sort by
AI chatbots and you: How AI chatbot behaviors affect user trust and performance
No full textThis thesis presents three projects centered on AI chatbots and trust, with the third extending to problem-solving performance. Chapter 2 examines user trust in AI chatbots that include and exclude citations for accurate and inaccurate responses. Results suggest that users are capable of properly calibrating their trust for appropriate reliance when given the opportunity, and citations do not influence their trust. Chapter 3 similarly examines the influence of citations in AI chatbot
responses, but within the context of politically charged topics where confirmation bias is common. Results showed that while citations again had no effect on trust, users reported similar levels of trust in confirming and moderate responses but reduced trust in challenging responses. This suggests that a chatbot presenting a balanced view can be equally trusted by users as a confirmatory chatbot. Finally, Chapter 4 examines how different AI chatbot behaviors, specifically question-asking frequency and communication style, affect user problem-solving performance and trust when the chatbot is used to help complete tasks. No significant main effects on user performance were discovered; however, users reported significantly higher trust in chatbots that were direct and did not ask questions. Notably, age and other individual differences had a greater impact on performance and trust than did the experimental factors. Together, these works and accompanying literature reviews on trust and problem solving help fill some of the gaps in current human-AI chatbot research, providing a stronger foundation for further study
Determining the Optimal Timing and Economic Return of Corn Fungicide Applications Using a Network Meta-Analysis
No full textA network meta-analysis was conducted to assess the efficacy of fungicides in reducing disease and protecting yield in corn. Uniform protocols were designed to test the efficacy of 12 widely available corn fungicides applied at one of the following timings: in-furrow with the seed at planting, applied 5.1 cm to the side and 5.1 cm below the seed at planting, 10 to 12 leaves with a visible collar, tasseling to silking (VT/R1), or milk stage. A total of 152 trials were conducted across 18 states in the United States and Ontario, Canada, from 2019 to 2022. Studies were analyzed using network meta-analyses to determine the fungicide efficacy and expected yield benefit of individual products compared with a nontreated control (NTC). All fungicides significantly reduced disease severity compared with the NTC (P < 0.001), and all fungicides resulted in greater yields compared with the NTC, except for Xyway LFR. Final disease severity influenced yield effect size, with fungicide application resulting in a greater yield effect size when final disease severity exceeded 5%. Fungicide application timing also influenced yield effect size, with fungicides applied at VT/R1 resulting in significantly lower disease (–7.6%) compared with the NTC. The yield effect size was typically greater in studies with the fungicide applied at VT/R1 compared with applications occurring at planting. Economic analyses concluded that expected net benefits were positive for all fungicides tested except for Delaro Complete and Xyway LFR. Most fungicides resulted in greater breakeven probabilities with increasing disease severity. The results emphasize that fungicide applications occurring at VT/R1 and when disease severity exceeds 5% are more likely to result in a positive economic gain.This article is published as Dangal, Nabin K., Maria Oros, Jason Lo, Isaac Baumann, Damon L. Smith, Tom W. Allen, Alyssa K. Betts et al. "Determining the Optimal Timing and Economic Return of Corn Fungicide Applications Using a Network Meta-Analysis." PhytoFrontiers (2025). doi: https://doi.org/10.1094/PHYTOFR-08-25-0079-RThis study was organized by the Corn Disease Working Group and conducted through collaboration across multiple states and provinces. Partial support for personnel conducting the work in individual states was provided by local checkoff organizations and agricultural research entities, including the Alabama Soybean Producers, the Arkansas Corn and Grain Sorghum Board, the Indiana Corn Marketing Council, the Kentucky Corn Growers Association, the Louisiana Soybean and Grain Board, the Corn Marketing Program of Michigan, Michigan AgBioResearch, and Grain Farmers of Ontario. Additional funding for field trials was provided by the USDA National Institute of Food and Agriculture through Hatch projects IND00162952 and 1025521 (Grant 2022-67013-37079) and IOW03098 and under the USDA Agricultural Research Service project number MIS-402050
The stunt of stunted silk: A novel pollination control mechanism in maize
No full textDear Editor,Existing pollination control strategies in plants can be widely classified into biological and nonbiological measures. The two main biological control measures are male sterility and gametophytic incompatibility systems (Kempe and Gils 2011). What if there is a genetic pollination control system with an application in “baby corn” breeding?This article is published Siddique I Aboobuckerصديق أبوبكر, Sidramappa C Talekar, Ursula K Frei, Bing Yang, Thomas Lübberstedt, The stunt of stunted silk: A novel pollination control mechanism in maize, Plant Physiology, Volume 200, Issue 1, January 2026, kiaf625, https://doi.org/10.1093/plphys/kiaf625National Science Foundation (NSF) Plant Genome Research Program award IOS-2428162 (S.I.A., T.L.), Iowa State University Plant Sciences Institute (T.L.), Iowa State University Crop Bioengineering Center (S.I.A., T.L.), Indian Council of Agricultural Research-National Agriculture Higher Education Policy-Institution Development Plan, awarded to the University of Agricultural Sciences, Dharwad, India (S.C.T.)
Optimal control for wave energy converters emphasizing low variance in power absorption
No full textWave energy conversion offers a high potential of renewable energy, with drawbacks in the cost of power production. Optimizing the control of wave energy conversion is critical in making use of this abundant resource and is responsible for advancements in obtaining maximum power absorbed for a given system. In addition to the quantity of energy produced, the quality must also be considered for ensuring consistent and high-quality power delivery suitable for long-term operational stability and integration into electrical grids. In this paper we formulate an optimal control solution with power quality introduced as a cost factor within the objective function. The foundation of this work is passed from (Abdulkadir & Abdelkhalik, 2022) where the power constrained bang-singular-bang control is used in comparison to our control solution. Results for the ‘Power Quality’ control show the trade-off that decreasing variance has on average power production. The Effective Firm Capacity (EFC) is used as an index for measuring usability of power which depends on the average power absorbed and the variance of the power. Optimal results for EFC were found to have a weighted parameter of \gamma=3 for given simulations, resulting in a 22% decrease in variance, a 7% decrease in average power and a 5% increase in EFC from the control case. The quality factor optimal control formulation shows the ability to think forward in the WEC energy problem and integrate end point variability into the control design for the system and improve the overall feasibility of WEC
Unified interactive frameworks for modular coding and circuit design
No full textThis thesis presents a unified framework for scalable, modular, and interactive development across software and hardware domains. We introduce BHDL, a declarative embedded language for PCB design that enables concise, hierarchical schematics and layouts within a Jupyter-powered environment. Building on this, we propose a deep reinforcement learning-guided Monte Carlo Tree Search approach for circuit routing, achieving flexible, high-performance automation adaptable to diverse design constraints. To overcome the limitations of traditional notebooks, CodePod extends Jupyter with hierarchical scoping and language-agnostic modularity, supporting large-scale, multi-language projects. Finally, Kernel-FFI enables seamless cross-language function calls and object referencing in interactive workflows, eliminating boilerplate and supporting object-oriented patterns and recursive remote function calls. Together, these contributions establish a cohesive ecosystem for interactive, modular, and scalable development in both coding and circuit design
Multispectral rheology
No full textRheology plays a pivotal role in characterizing the flow and deformation behavior of materials, underpinning advances across disciplines such as biology, materials science, and industrial processing. This dissertation presents a comprehensive investigation of contemporary rheological tools - conventional rheometers, Dynamic Light Scattering (DLS), Optical Tweezers (OT), and Laser Speckle Rheology (LSR) - to probe the visco-elastic mechanisms of composite hydrogels designed as brain-tissue mimics and polymer-based contact lenses formulated in-house.
Despite extensive literature on soft hydrogels and biomaterials, no prior studies have documented the rheological response of hydrogel-based tissue mimics under high strain rate events such as shock wave exposure. Similarly, while drug-infused contact lenses are widely commercialized, their drug release behavior has rarely been correlated with the underlying polymer relaxation dynamics that govern lens performance. Conventional approaches, including rotational rheometry and Atomic Force Microscopy (AFM), are often intrusive and limited in dynamic range, whereas optical methods have historically been dismissed due to tissue opacity [1].
This work challenges those limitations by integrating optical and mechanical rheometry within a unified framework. Each technique – mechanical, scattering based, and speckle-based - is evaluated in terms of measurement scale, sensitivity, operational conditions, and material applicability. The central hypothesis is that multi-spectral rheological analysis, combining optical and mechanical modalities, provides a more comprehensive understanding of material behavior across frequency and opacity domains.
To test this hypothesis, a custom OT system was assembled and calibrated using water at room temperature. Calibration revealed a critical factor influencing trapping accuracy: near-wall hydrodynamic drag caused by sample confinement. Although the OT setup was unable to probe stiffer materials representative of brain-tissue mimics, this limitation motivated the adoption of DLS, which successfully characterized stiffer, transparent composites modified with plasticizers. However, DLS proved inadequate for highly elastic or opaque samples, prompting the development of an LSR system. LSR measurements successfully extended rheological characterization to both opaque and rigid hydrogels, overcoming prior constraints.
Together, these complementary techniques establish a multi-spectral, multi-frequency rheological framework that bridges optical and mechanical methods. This dissertation thus serves as a methodological and analytical reference for selecting appropriate rheological tools based on material transparency, stiffness, and experimental conditions. Furthermore, the high strain-rate investigations presented here pave the way for future studies on other organ-mimicking hydrogels (e.g., liver and lung), contributing to the development of experimental datasets crucial for validating computational models of dynamic tissue behavior
Semi-Inducibility of some small graphs
No full textLet H be a fixed graph whose edges are colored red and blue and let β ∈ [0, 1]. Let I(H, β) be the (asymptotically normalized) maximum number of copies of H in a large red/blue edge-
colored complete graph G, where the density of red edges in G is β. This refines the problem of determining the semi-inducibility of H, which is itself a generalization of the classical question of determining the inducibility of H. The function I(H, β) for β ∈ [0, 1] was not known for any graph H on more than three vertices, except when H is a monochromatic clique (Kruskal-Katona) or a monochromatic star (Reiher-Wagner). We obtain sharp results for some four and five vertex graphs, addressing several recent questions posed by various authors. We also obtain some general results for trees and stars. Many open problems remain.This preprint is from Balogh, József, Bernard Lidický, Dhruv Mubayi, Florian Pfender, and Jan Volec. "Semi-Inducibility of some small graphs." arXiv preprint arXiv:2601.03433 (2026).
https://doi.org/10.48550/arXiv.2601.0343
Skeletal muscle health in disease conditions: Captopril effects on inflammation and autophagy in Zucker diabetic fatty rats, and imoxin effects on inflammation, ER stress, and muscle function in dystrophic mdx mice
No full textSkeletal muscle plays a critical role in physical activity and systemic homeostasis. However, its structure and function are frequently compromised in both metabolic and genetic diseases. This dissertation investigated the effects of long-term and systemically delivered pharmacological interventions on skeletal muscle to preserve muscle health in two disease models: type 2 diabetes mellitus (T2DM) and Duchenne muscular dystrophy (DMD).
In the first study, the effect of chronic angiotensin-converting enzyme (ACE) inhibition using captopril was examined in the soleus and extensor digitorum longus (EDL) muscles of 14-week-old Zucker diabetic fatty (ZDF) rats, a model of T2DM. Despite reduced muscle mass and systemic metabolic dysfunctions, captopril did not rescue muscle weight, inflammatory signaling, autophagy markers, or protein synthesis pathways in either oxidative or glycolytic muscles. Notably, the skeletal muscle of 14-week-old ZDF rats did not exhibit impaired inflammation or autophagy, and protein synthesis pathways although muscle mass and AMPK signaling were decreased.
Building on this, we next examined a more severe muscle-specific pathology in the mdx mouse model of DMD. In the second study, we confirmed that long-term prednisolone (Pred) treatment improved specific force production and reduced disease-induced inflammatory responses in dystrophic diaphragm. We newly discovered that Pred decreased endoplasmic reticulum (ER) stress response, and apoptosis in the dystrophic diaphragm, highlighting its efficacy beyond anti-inflammation.
The third and fourth studies investigated the therapeutic potential of imoxin (IMX), a specific protein kinase R (PKR) inhibitor, in dystrophic skeletal muscles (diaphragm, hindlimb muscles). While IMX slightly reduced fibrosis in the diaphragm examined by trichrome staining, it did not suppress increased PKR activation (phosphorylation), inflammatory signaling, or ER stress in the diaphragm. Muscle force generation remained impaired in the diaphragm, and IMX had no restorative effects. In the fourth study, EDL showed increased fibrosis and decreased specific force but soleus specific force showed similar between C57-Vehicle and mdx-Vehicle. While EDL showed increased vulnerability in repetitive eccentric contractions by disease and increased passive force, soleus showed similar fatigue resistance and passive force among groups. IMX delivered no efficacy in terms of force production in both soleus and EDL. Biochemical assay in the gastrocnemius of mdx mice exhibited continuous elevation of inflammation and ER stress with no increase and inhibition of PKR, indicating that dystrophic muscle pathology could be independent of PKR activation.
Together, these findings demonstrate that captopril and PKR inhibition through imoxin did not reverse skeletal muscle pathophysiology in muscles of metabolic or dystrophic models highlighting the challenge of effectively targeting PKR and complexity of regulatory mechanism of PKR in dystrophic muscle. However, long-term glucocorticoid treatment effectively restored cellular dysfunction, particularly alleviated increased level of ER stress and apoptosis in dystrophic diaphragm. These results provide insights into the context-dependent effect of existing pharmacological interventions in the muscle of chronic metabolic and genetic diseases, and our data indicates that PKR is unlikely to be a dominant regulator of inflammation or ER stress in dystrophic limb muscles
Investigations focused on understanding the biological features preceding pelvic organ prolapse in sows
No full textTo increase the sustainability of the U.S. swine herd, it is critical to maximize sow reproductive efficiency. While significant improvements have been made in several areas of production efficiency, including reproductive performance, over the past decade, sow mortality rates have increased, with a disproportionate increase due to pelvic organ prolapse (POP). While the cause of this anatomical phenomenon remains unknown, POP typically occurs during the peripartum period and contributes to nearly 25% of all sow mortality. A previously established perineal scoring (PS) system during late gestation based on phenotypic observations correlated with POP risk was utilized. This PS system was then applied as a research tool to better understand multiple facets of complex biological systems, which may contribute to POP risk and outcome in sows. Fecal and vaginal swabs, along with blood, were collected from sows at high and low risk for POP. Fecal microbiota was evaluated using 16S rRNA gene sequencing, and fecal and vaginal microbial communities were compared (Chapter 3). Differences in fecal microbiota were discovered between sows at various POP risk levels. Vaginal swabs were additionally utilized for metagenomic shotgun sequencing, and the functional potential of sow vaginal microbiota was evaluated in relation to POP (Chapter 4). As a result of microbiota discoveries, a mitigation strategy was developed using vaginal infusions of an antibiotic to reduce POP incidence (Chapter 5). Serum analytes, trace minerals, and potential biomarkers were evaluated, and differences between sows at various POP risk and outcome were discovered (Chapter 6). Collectively, these data suggest differences exist between microbial populations, serum analytes, trace minerals, and biomarkers between sows at various risks for POP during late gestation. This dissertation provides a more comprehensive understanding of the biological features associated with increased POP risk in sows to enable development of industry focused mitigation strategies
Effect of external stimulation on neural stem cell behavior
No full textNeurodegenerative diseases and brain injury often result in significant neuronal loss and cognitive deficits. Unfortunately, the mammalian central nervous system (CNS) has limited innate repair mechanisms to combat the damage caused by neurodegenerative diseases. Cell-based therapies, such as neural stem cell (NSC) transplantation, are a promising strategy for replacing lost or damaged cells, current methods are not yet ready for clinical use. The limitations of current NSC transplantation strategies include NSCs within the transplanted cell population being at differing states of differentiation or maturation or lack the presence of the cell type of interest. Finding a differentiation protocol that would be scalable, cost-effective, and efficient has been a focus for researchers developing tissue and cellular engineering strategies. Recently, utilizing external stimulation such as electrical stimulation (ES) or magnetic stimulation (MS) has become a promising method for guiding the differentiation of NSCs. The objective of this project was to investigate the application of ES and MS to enhance the differentiation of NSCs in a parameter-dependent manner for use in future nervous system regeneration or repair strategies.
In this work, adult hippocampal progenitor cells (AHPCs) were used to evaluate the effects of ES and MS on modulating cell proliferation and differentiation in vitro. Our work revealed that both ES and MS can be used effectively to guide the differentiation of AHPCs into specific cell types in a parameter-dependent manner while still maintaining high cell viability. Both stimulation methods resulted in enhanced differentiation of the AHPCs into both neuronal and oligodendrocyte cell types, with oligodendrocyte differentiation being the most highly enhanced. We were able to achieve this enhanced differentiation after only 7 to 10 days of ES or MS, respectively. These results showed that ES and MS have the potential to be used as a cost-effective and efficient strategy for selective AHPC differentiation.
Additionally, in collaboration with engineers our group has worked on developing a stem cell culturing platform for the study of various neurological disorders in vitro for a potential drug screening tool. We were able to successfully culture AHPCs as three-dimensional non-adherent neurospheres within individual chambers in a novel microfluidic device. The AHPC neurospheres remained viable within the devices and were able to proliferate with observable neurosphere growth over time. The AHPCs within the neurospheres also retained their ability to differentiate into multiple neural cell types within the devices. Given the ability to culture multiple culture multiple neurospheres within a single device, this cell culture platform would be promising platform as an in vitro screening tool to evaluate the effects of therapeutic strategies in a highly efficient manner.
Overall, this work provides evidence that using external stimulation techniques such as ES and MS can be effective strategies for guiding the differentiation of NSCs. Using these differentiation strategies in combination with microfluidic cell culture platforms may lead to the development of more effective strategies for CNS rescue or repair