16144 research outputs found
Sort by
ParDcuhre: A Scalable Parallel Solution for Multivariate Integration with CUDA
Full text linkMathematical integration is a common process in a wide array of fields such as medical imaging, statistical analysis, orbital mechanics, and simulation modeling. Since analytical solutions to many integration problems are impossible to obtain, they are instead numerically approximated. There exists a variety of numerical integration software used to estimate such integrals; one such software is DCUHRE which employs the global adaptive algorithm over a hyperrectangular region to estimate a given function. However, as the integral dimension increases, the number of region evaluation points increases exponentially. While DCUHRE was written to accommodate parallel execution on multiple processors (subregion-level parallelization), this level of parallelization is significantly less than what can be achieved using GPUs, thus resulting in significantly longer execution times in higher dimensions. Our parallel solution ParDCUHRE implements evaluation-point-level parallelization and achieves speedups well into the triple digits with selected integrals (with dimensionality as high as 25) while yielding results comparable to the original DCUHRE
Experimental Validation of Bio-Inspired Control: 4 Story Shear Structure
Full text linkWith the rise of rapid urbanization, skyscrapers are increasingly common, making them vulnerable to environmental pressures such as high winds and earthquakes. These forces threaten the structural stability of tall buildings, necessitating the development of active control methods to mitigate their effects and maintain structural integrity. Traditional control systems utilize numerous sensing nodes that feed data into a centralized system, which then determines the appropriate actions for the actuators. However, this centralized approach can introduce substantial lag due to the overwhelming amount of data being processed. By transitioning to a decentralized wireless system, where sensors directly feed data to control nodes, scalability is improved, and lag is reduced. However, the wireless nature increases the risk of data loss. The proposed solution draws inspiration from the biological central nervous system, with sensing nodes performing upfront signal processing. A weighting matrix determines how data is transmitted to motor neurons, which decide actuator firing and force application. Using the Linear Quadratic Regulator (LQR) control theory, weights are optimized by modeling structural behavior while striving to minimize floor displacement and acceleration. The experimental setup involved a four-story shear structure outfitted with position sensors and accelerometers, which fed data into sensor nodes. These nodes relayed information to control nodes that calculated the necessary control force, which was then transmitted to actuator carts, ensuring effective stabilization of the structure
Investigation of Tc Change Due to Proton Irradiation in TBCCO Superconductors
Full text linkWe studied the effect of 600 keV proton irradiation on thin film cuprate superconductors. A 650 nm thick TBCCO-2212 sample was subjected to a series of proton irradiations totaling a fluence of 5.2 x 10^16 protons/cm^2. Superconducting critical temperature (Tc) was decreased from 100 K towards zero Kelvin, and the normal state resistivity increased accordingly. The rate of Tc reduction to resistivity increase will be used to discuss the fundamental properties of the superconductor
Belonging Among LGBTQ+ Students at Hope College
Full text linkAt Christian Colleges and Universities (CCUs), studies estimate that approximately 12% of a given college\u27s student population identifies as LGBTQ+ (REAP, 2021). Unfortunately, research literature also indicates that LGBTQ+ college students experience significant challenges, including, but not limited to, discrimination from peers and staff, alienation from the wider campus environment, and inhospitable housing conditions (REAP, 2021; Rankin et al. 2019; Kortegast, 2017; Craig et al. 2017; Wolff et al. 2016, Christie, 2021). Academic research on LGBTQ+ students at CCUs is somewhat limited, even more so research that uses a qualitative orientation, attempting to gain in-depth insights into LGBTQ+ students\u27 experiences. This current study involved a sample of seven LGBTQ+ Hope College students, who participated in six weeks of focus groups and documentary photography, using a participatory action method of research called photovoice. The photovoice method enables participants to record and reflect upon their personal and community strengths and concerns, promote critical dialogue through group discussions, and influence policymakers through data. This study aimed to understand how LGBTQ+ students at one CCU experience belonging, and the lack thereof, within the campus community
Prism Focus Exhibition
Full text linkA poster advertising a focus exhibition of artwork selected by Hope College\u27s Prism student organization. The exhibition was held from January 14-March 1, 2025.https://digitalcommons.hope.edu/kam_poster/1068/thumbnail.jp
The Taxonomy of Evidence-Based Practices for United States History Education
Full text linkTeaching the next generation of leaders, thinkers, and citizens is the one of the primary roles for Social Studies teachers. This can be especially true for those instructors who teach United States history since this class focuses on the past and future of our country. To help in this teaching endeavor, this study has created a Taxonomy of Social Studies Evidence Based Practices. A systematic and exhaustive review of the literature was completed finding 17 articles in which the taxonomy was created. Within this taxonomy, arms related to (1) effective technology use, (2) place-base-learning, (3) primary documentation sourcing, and (4) culturally relevant teaching practice will be emphasized. This poster will discuss this taxonomy in depth and how Social Studies teachers can improve their teaching using these important Evidence Based Practices
The Effects of Visual Dominance and Object Naming on Word Learning in Infants
Full text linkChildren learn new words every day, a majority of which are nouns (e.g., car, dog, ball). This raises the question: how do infants connect the correct words to each object? Previous research suggests that infants as young as 12 months can learn object names by tracking which words and objects appear together (Smith & Yu, 2008). Additionally, object names may be learned better when the object being named is visually dominant, meaning larger in the infant\u27s view (Yu & Smith, 2012). However, this correlation has not been tested experimentally. This study tests whether visual dominance enhances word learning in infants. Thirty-two infants (17 females; M age = 12.9 months, SD = 4.1) participated in a standard object-name learning task similar to that of Smith and Yu (2008). However, in our task, infants viewed on a screen four target objects that we modified to be presented as larger (dominant) than eight distractor objects. In each trial, a target object was presented along with two distractor objects. In the Naming condition, target objects were named; in the No-Naming control group, they were not. After 160 training trials, word learning was assessed by how long infants looked at the correct target object when it was named, using eye-tracking. If learning occurs, we expect infants to look longer at the named target object in the Naming condition. Results showed that infants did not look longer at named target objects in the Naming condition compared to the No-Naming control. This suggests that, even though target objects were visually dominant, infants were not learning the object names. A follow-up study is underway with equally-sized objects to further explore the role of visual dominance in word learning and will compare effects of visual dominance in both conditions
Telling Their Stories: Student-Curated Focus Exhibitions
No full textA poster advertising a focus exhibition of artwork selected by Hope College students. The exhibition was held from August 29-December 13, 2025 and featured student personal responses to artworks.https://digitalcommons.hope.edu/kam_poster/1070/thumbnail.jp
The Anchor: March 2025, Issue Three
Full text linkThe Anchor began in 1887 and was first issued weekly in 1914. Covering national and campus news alike, Hope College’s student-run newspaper has grown over the years to encompass over two-dozen editors, reporters, and staff. For much of The Anchor\u27s history, the latest issue was distributed across campus each Wednesday throughout the academic school year (with few exceptions). In recent years The Anchor moved to monthly print issues and a more frequently updated website. The Anchor is now published in print twice per academic semester. Occasionally, the volume and/or issue numbering is irregular
Self-Scheduled Category Learning in College Students
Full text linkLearning new information can be a complex, gradual process. It is important to understand how learning plays out over time. Research has found that spaced schedules — studying new information in multiple sessions spread out over time — are more effective for learning than massed schedules — studying everything in one long session (Carvalho & Goldstone, 2013; Cepeda et al., 2006, Kornell & Bjork, 2008). However, when people schedule their own learning, they prefer massing (Kornell & Bjork, 2008; Tauber et al., 2013). Recent research suggests another type of schedule, clustering, might be more effective than both massing and spacing. Clustering involves spacing out clumps of massed studying (Slone et al., 2022; Yan et al., 2017). Little research has examined whether people choose to cluster when studying new information. The current study aims to fill this gap by examining how often people cluster their studying when learning new categories, and comparing the effectiveness to massing and spacing. Participants will see exemplars from nine categories (or families ) of alien creatures. Importantly, on each trial participants will choose which family to study next. We will examine which schedule (spaced, massed, or clustered) each participant implemented for learning each family. Participants will then be tested on their ability to correctly classify new creatures. We will examine which schedule(s) produce the best learning. We hypothesize that categories studied with clustered schedules will be learned best because clustering helps learners identify both similarities among family members (when viewed back-to-back as in massing) as well as differences between families (when switching categories as in spacing). Finally, participants will be asked to indicate their preferred learning schedule. We predict that the majority of participants will indicate that massing and clustering are equally effective. Our study has HSRB approval, and we are currently programming the experiment. We will begin collecting pilot data later in the semester