33760 research outputs found
Sort by
Evaluating Early-Life Behavioral Responses to Social Cues in Cichlid Fish
No full textSocial behavior can vary across development stages. An animal may prefer to affiliate with a protective parent, groups of peers, or be solitary depending on its age and needs. Cichlid fish are powerful models to study social behavior due to their robust sociality, rapid diversification and genetic tractability. Many mature social behaviors and signaling pathways have been characterized in the cichlid Astatotilapia burtoni. Juvenile A. burtoni (fry) behavior is less understood. Studying fry behavior elucidates early social affiliations, granting ecological significance for survival and social organization as well as a model for the evolution and mechanisms of early behavior. We present fry with social stimuli in a two-choice assay, allowing chemosensory and visual perception to determine if they seek their mother or siblings (kin). We expect fry to prefer their mother or kin over another choice, determined by time spent close to one stimulus over the other. Testing n=4 trials for preference of mother vs. male, our results indicate overall that fry do not show a clear preference for their mother. However, preference shifts with age suggest an onset window for maternal preference with neurosensory maturation. Testing n=6 trials of kin versus non-kin, n=6 trials of kin vs. heterospecifics, and n=2 trials of large vs. small kin groups, our results indicate that fry seek peer groups, but cannot discern their kin from other fry. We propose a model that particular social preferences shift with fry development stages. Further research may test preference onsets with age, determine the influential sensory modalities for fry behavior and their associated neural activity patterns.This work was supported in part by grants to Scott A. Juntti from the National Institutes of Health (R35GM142872); the National Science Foundation (IOS-1825723); the Human Frontiers in Science Program (RGY0079-2018); and a fellowship to Coltan G. Parker from the National Science Foundation (DBI-2209257)
Large-Scale Behavior of Some Stochastic PDES
No full textIn this dissertation, we include four projects that study the large-scale behavior of some stochastic PDEs. The stochastic PDEs that we study are the Kardar-Parisi-Zhang (KPZ) equation and the stochastic heat equation (SHE).
The Kardar-Parisi-Zhang equation is a nonlinear stochastic PDE, known as the default model for random interface growth in physics. To study the Kardar-Parisi-Zhang equation, we use the approach of Hopf-Cole transform, which relates the KPZ equation to the stochastic heat equation. In spatial dimension one, the KPZ equation and the stochastic heat equation have received extensive studies, but the equations that are boundary driven or in higher dimensions remain more mysterious.
In the first part of this dissertation, we consider the stochastic heat equation and KPZ equation in spatial dimension two, which is a critical dimension, and the solutions can only be made sense as scaling limits. In the first project, we consider a nonlinear version of the stochastic heat equation and prove its limiting second-order Gaussian fluctuation. This result has been published in paper [1]. In the second project, we show a mesoscopic averaging phenomenon for the local averages of the two-dimensional KPZ equation. This result has been published in [2].
The second major part of this dissertation is a project that considers the KPZ equation in a one-dimensional half-space domain with a Neumann boundary condition. This half-space KPZ model exhibits a “depinning” phase transition as the boundary condition changes. We study the half-space KPZ equation starting from stationary Brownian initial data. We obtain its optimal fluctuation exponents in both the subcritical and critical regimes of the phase transition, and give an optimal upper bound for the fluctuation exponents in the supercritical regime. We also compute the average growth rate as a function of the boundary parameter. This result has been submitted for publication in paper [3].
In the last part of this dissertation, we also include a work that studies the time-dependent spatial averages of a long-range critically correlated stochastic heat equation in spatial dimension three or higher. These spatial averages have different limits under different space-time scales. This result has been included in preprint [4]
In The Concise Evolutionary Essays, Robert B. Graber, ed.
No full texthttps://www.berghahnbooks.com/title/GraberCarneir
Solid-State Battery Database Mastersheet
No full textThe data in this database was extracted from published solid-state battery papers. For more information on the data extraction methods please contact [email protected] thousands of annual publications on battery devices, standard reporting practices are distinctly absent, resulting in a wide variation of reported data that is difficult to search and analyze. To address this, we constructed a database of research-level solid-state batteries with a preliminary ontological framework reflected in 167 columns and populated with 245 batteries resulting in over 40,000 data-points. This project’s significance is threefold: (1) Our variables include materials, components, fabrication, and cell testing, whose combination is essential to describe and properly assess a solid-state battery. (2) Our inclusion of “not applicable” and “not reported” values allow us to analyze reporting practices and gain insights on ”missing data” in battery publications. (3) We present a structure for battery data with well-defined variables, facilitating formatted reporting. Key applications include a streamlined visualization and search workflow employing numerical and categorical criteria, quantitative analysis of reporting practices, a training set for AI paper-data extraction, and connections with existing battery ontologies.This work was supported by the U.S. – Israel Energy Center program managed by the U.S. – Israel Binational Industrial Research and Development (BIRD) Foundatio
Impact of 2-isopropylmalate synthase (LeuA) Knockout on Escherichia coli Growth and Bacteriophage Replication
No full textWith antibiotic resistance on the rise, researchers are looking at phage therapy to fight antibiotic-resistant bacteria. Bacteriophages target and hijack bacterial cells, using their macromolecules to replicate themselves and ultimately kill the host cell. While the function of bacteriophages is understood, researchers are still unaware of how certain metabolic pathways are manipulated during bacteriophage replication. In our research, we explore how the knockout of the leuA gene, which is crucial to the leucine synthesis pathway in E. coli, affects bacteriophage growth. We compared the growth and phage susceptibility of a leuA knockout E. coli strain to its wild-type parent by generating growth curves, lysis curves, and phage titers under varying leucine conditions. Data were collected using spectrophotometry, plaque assays, and time-point phage quantification, and analyzed to determine how leucine availability and leuA deletion affect bacterial fitness and phage replication. Our growth curves showed that the knockout grew slightly better than the parent strain in LB media, which contains leucine, but both strains grew similarly in M9 media, which does not contain leucine. The lysis curves, which essentially measure bacteriophage replication, showed almost no difference between the parent and knockout strains, indicating that bacteriophage replication is not affected by the LeuA gene. Our plaque assay plates also showed similar levels of bacteriophage growth, corroborating our results for the growth and lysis curves. However, our two-time point titer test showed that bacteriophage grew significantly better in the parent strain than in the knockout strain, which contradicts our data for the other experiments we conducted. That being said, it may also suggest a role leucine has in phage replication. Our results demonstrate that knockout of the LeuA gene has minimal to no impact on the L-Leucine biosynthesis pathway within E. coli, and subsequently E. coli and bacteriophage growth. Future work will involve the knockout of additional genes within the L-Leucine biosynthesis pathway to assess whether or not the observed effects of LeuA are unique. We will also repeat previous experiments to address potential error, and ensure redundancy
Age-related temporal processing deficits: The relationship between gap detection and temporal cue discrimination
No full textTemporal processing abilities are crucial for encoding auditory information. These abilities decline with age, particularly impacting the ability to perceive brief temporal cues. Furthermore, age-related temporal processing deficits appear to be particularly detrimental for speech processing with degraded auditory input, such as that experienced by cochlear-implant users. The goal of this study was to investigate how aging affects performance on different auditory temporal processing tasks in individuals with normal hearing presented simulated electric hearing. First, we hypothesized that reducing spectral information and increasing temporal complexity in a sound would negatively impact temporal processing, with older listeners showing greater deficits. The listeners completed a gap detection task with four stimuli varying in temporal complexity and spectral information. Gap detection thresholds were influenced by both spectral and temporal characteristics of the stimuli, as well as by the listeners’ age although no significant interaction between condition and age group were observed. Second, we hypothesized that reducing spectral information would affect listeners’ ability to discriminate between temporal speech cues, with older listeners showing more difficulty. Listeners completed a temporal cue discrimination task where they identified differences in silent interval durations using the word pair “Dish” and “Ditch” presented on a 7-step continuum, ranging from 0-60 ms of silence. Stimuli were presented in both unprocessed and vocoded conditions. Slope values of the psychometric functions for the vocoded stimuli were shallower compared to unprocessed speech, demonstrating that reduced spectral information made the task more difficult and increased reliance on temporal cues. In addition, no significant effects of age group across conditions were observed, suggesting that age did not impact performance on the temporal cue categorization task. Finally, we hypothesized that the individual variability in speech discrimination performance would correlate with gap detection thresholds. No significant correlations were found between performance on gap detection and temporal cue discrimination tasks, suggesting that these tasks may rely on different auditory, linguistic, or cognitive mechanisms. Overall, these findings highlight the importance of temporal processing for understanding speech with degraded spectral input and raise important considerations for older cochlear-implant users. Addressing temporal processing deficits through improvements in device programming may help audiologists enhance speech perception and listening experiences for individuals with cochlear implants
Supplementary material for Applying Wearable Sensors and Machine Learning to the Diagnostic Challenge of Distinguishing Parkinson's Disease from Other Forms of Parkinsonism
No full textParkinson's Disease (PD) and other forms of parkinsonism share motor symptoms, including tremor, bradykinesia, and rigidity. This overlap in the clinical presentation creates a diagnostic challenge, underscoring the need for objective differentiation. However, applying machine learning (ML) to clinical datasets faces challenges such as imbalanced class distributions, small sample sizes for non-PD parkinsonism, and heterogeneity within the non-PD group. This study analyzed wearable sensor data from 260 PD participants and 18 individuals with etiologically diverse forms of non-PD parkinsonism during clinical mobility tasks, using a single sensor placed on the lower-back. We evaluated the performance of ML models in distinguishing these two groups and identified the most informative mobility tasks for classification. Additionally, we examined clinical characteristics of misclassified participants and presented case studies of common challenges in clinical practice, including diagnostic uncertainty at the initial visit and changes in diagnosis over time. We also suggested potential steps to address dataset challenges which limited the models' performance. We demonstrate that ML-based analysis is a promising approach for distinguishing idiopathic PD from non-PD parkinsonism, though its accuracy remains below that of expert clinicians. Using the Timed Up and Go test as a single mobility task outperformed the use of all tasks combined, achieving a balanced accuracy of 78.2%. We also identified differences in some clinical scores between participants correctly and falsely classified by our models. These findings demonstrate the feasibility of using ML and wearable sensors for differentiating PD from other parkinsonian disorders, addressing key challenges in diagnosis, and streamlining diagnostic workflows.Funding was provided by a University of Maryland MPower Seed Grant Award (R.v.C. and M.P.C), the Rosalyn Newman Foundation (L.M.S), and the University of Maryland Claude D. Pepper Older Americans Independence Center (P30-AG028747; R.v.C)
Active Seismic Exploration of Planetary Subsurfaces via Compressive Sensing
No full textThe software supports the following study:
We present a method for improving seismic data collection on planetary surfaces such as the Moon and Mars. This approach is based on recent advances in compressive sensing technology to reduce the number of data collection points required compared to conventional methods without sacrificing the quality of the resulting subsurface images. We demonstrate its effectiveness using both synthetic and field data from locations with similarities to planetary surface environments. The method is then applied to reanalyze seismic data collected by the crew of the Apollo 14 and 16 missions. Our study has implications for mission planning, as this method can make space missions more efficient by reducing the equipment and time to collect geophysical data on planetary surfaces. It also makes it possible to reconstruct missing or damaged data, improving the quality of imagery and enhancing our understanding of the interior of other worlds.https://doi.org/10.1029/2024EA00382
SMALL AND LARGE PERCEPTION MODELS FOR ROBOTIC NAVIGATION
No full textComputer vision is fundamental to advancing robotic navigation and autonomous driving systems, enabling machines to interpret visual data critical for interaction with complex and diverse real-world environments. We address many critical challenges in these domains by developing advanced vision methods. Our approaches leverage the complementary strengths of small-scale specialized models, and large-scale vision-language models (LVLMs) with better generalized zero-shot capabilities. Specifically, small-scale models typically focus on specialized tasks, such as object detection, segmentation, and terrain classification. On the other hand, large-scale vision models, particularly vision-language models (VLMs), leverage extensive training data to capture richer contextual information and enhance generalization. Yet, these benefits often come at the expense of increased computational demands and latency, and VLMs can still fail when confronted with complex or nuanced scenarios specific to certain tasks.To enhance perception capabilities in challenging scenarios, we introduce GA-Nav, an efficient transformer-based terrain segmentation approach explicitly designed for off-road robotic navigation. Our method simplifies the semantic segmentation task, emphasizing distinct terrain types to enhance navigation capabilities in unstructured outdoor environments. Additionally, we propose M3DETR, the first unified transformer-based architecture for 3D object detection in autonomous driving contexts, simultaneously modeling multirepresentation, multi-scale, and mutual-relations with transformers. To understand hallucinations and further improve model generalization and adaptability for navigation applications, we propose HallusionBench and AutoHallusion, the first few benchmarks designed to diagnose different types of hallucinations and systematically scale hallucination examples. In HallusionBench, we systematically analyze and benchmark accuracy, sycophancy, robustness, and various failure modes of existing LVLMs. We also propose AutoHallusion, the first fully automated pipeline capable of generating challenging hallucination cases based on the pattern we discovered. Leveraging insights from these benchmarks on VLM failure modes and hallucination patterns, we introduce ZSORN, advancing vision-language models for indoor zero-shot object navigation (ZSON). ZSORN provides a practical, retrieval-based solution for natural language-guided navigation, compared to traditional reinforcement learning approaches. Furthermore, we present CrossLoc3D and AGL-Net, robust global localization techniques that ensure reliable long-horizon navigation by using different map modalities, thereby providing enhanced robustness in GPS-denied regions. Through these advancements, we bridge the theoretical innovations in computer vision with practical, scalable solutions for sophisticated robotic and autonomous navigation systems
AN EXPLORATORY STUDY OF HOW PRINCIPAL SUPERVISORS SUPPORT PRINCIPALS’ INSTRUCTIONAL LEADERSHIP CAPACITY
No full textThis study investigated the role of principal supervisors in educational leadership within the K–12 Grand Royal School System (GRSS). It examined what principal supervisors do to build principals’ instructional leadership capacity—and to what extent their actions support principals in leading teaching and learning. Using a mixed-methods approach, including surveys, interviews, observations, and document analysis, the study explored the daily practices of principal supervisors and their interactions with principals.Findings show that GRSS principal supervisors value instructional leadership and often use goal-setting and formal observations to build principals’ capacity. However, their daily work is frequently consumed by operational demands, community concerns, and compliance tasks, which limit their ability to focus on instruction. Although most participants agreed that principal supervisors were accomplished instructional leaders, neither principals nor supervisors cited specific examples of support that led to instructional improvements. This gap was attributed to systemic barriers such as competing district priorities, ambiguous role expectations, and limited time.
In addition to these core findings, interviews revealed several unexpected insights, including supervisors’ lack of control over their schedules, inconsistent implementation of site visits, and a shared sense of nostalgia for earlier support structures. Supervisors also described the emotional toll of the role, marked by stress and strain from balancing competing demands without sufficient authority or systemic support. These perspectives provided valuable context for interpreting survey results and underscored the disconnect between the role’s intended purpose and its day-to-day realities.
While participants believed in the potential of the supervisor role to enhance leadership and improve learning outcomes, they acknowledged that current conditions often prevent it from functioning as intended. The study concludes with recommendations for restructuring the principal supervisor role and strengthening support for central office leaders who serve in this capacity