Northeastern University

NSF Awards · FY 2024 · 2024-09

A major challenge in deaf science education is the lack of standard signs in American Sign Language (ASL) for many scientific concepts. For example, one student might fingerspell a term, another might use a sign they created, and a third might use a different sign altogether. This variation can make it difficult for students to engage effectively in class without a shared understanding of scientific terminology. Collaborative problem-solving activities are known to improve the understanding of complex concepts, but traditional support methods mainly benefit hearing students. This makes it more challenging for deaf students who use sign language to participate fully. Additionally, deaf individuals are significantly underrepresented in scientific fields, which adds to their marginalization. To address these issues, the project will develop a new artificial intelligence tool designed to revolutionize collaborative learning for deaf students in science, helping them to better understand and communicate in university biology classes. The tool will use augmented reality, signed animations, and sign recognition to provide real-time information about the signs used in classroom conversations. The primary hypothesis of the research is that artificial intelligence-driven technology can significantly improve the collaborative experience and learning outcomes for deaf students. The project focuses on establishing common ground, which is particularly challenging in science courses where standard ASL signs are lacking. The team uses augmented reality to visualize scientific lexicon representations, including signing avatars and English captions. These aids complement existing learning strategies, such as parallel visual processing and the creation of new terms. This project will assist students in learning new terminology introduced by teachers or emerging from classroom conversations. It caters to the diverse needs of the deaf community in terms of language fluency, hearing ability, and use of assistive technologies by providing flexible, non-invasive learning supports. In support of the project goals, the team will convene co-design sessions, conduct prototype testing, and implement an experimental study to assess the impact of the tool. The project team includes experts in ASL scientific lexicons, learning sciences, human-computer interaction, and artificial intelligence. The goal is to improve inclusive education strategies, focusing on collaborative learning in science. The project contributes to human-computer interaction by identifying design principles for intelligent support to signing learners. It advances artificial intelligence through state-of-the-art sign recognition and generation systems, adaptive to learner variability, and incorporating facial expressions and prosodic features. In learning science, the project explores the relationship between adaptive scaffolds for lexical alignment, collaborative processes, and learning outcomes. In terms of deaf education, the project develops interventions supporting collaborative learning among deaf students. Acknowledging the diverse experiences within North American deaf communities, the initiative works to understand these nuances. If successful, the technology could generalize to other learning scenarios involving collaborating deaf students. This work will also support professional and scientific opportunities for deaf scientists, students, and trainees. This project is funded by the Research on Innovative Technologies for Enhanced Learning (RITEL) program that supports early-stage exploratory research in emerging technologies for teaching and learning. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

Strengthening American Infrastructure (SAI) is an NSF Program seeking to stimulate human-centered fundamental and potentially transformative research that strengthens America’s infrastructure. Effective infrastructure provides a strong foundation for socioeconomic vitality and broad quality of life improvement. Strong, reliable, and effective infrastructure spurs private-sector innovation, grows the economy, creates jobs, makes public-sector service provision more efficient, strengthens communities, promotes equal opportunity, protects the natural environment, enhances national security, and fuels American leadership. To achieve these goals requires expertise from across the science and engineering disciplines. SAI focuses on how knowledge of human reasoning and decision-making, governance, and social and cultural processes enables the building and maintenance of effective infrastructure that improves lives and society and builds on advances in technology and engineering. Amidst the global shift toward green and sustainable urban transportation systems, substantial investments have been made in infrastructure to facilitate the adoption of electric vehicles (EVs) in the United States. The placement of an EV charging station (EVCS) potentially has significant broader impacts on peoples' mobility, activity patterns, and visitation to nearby businesses during charging sessions. This provides an opportunity for policy makers to support local businesses (e.g., cafes, restaurants, grocery stores), particularly small and medium-sized enterprises, which play a pivotal role in maintaining community health, especially in vulnerable communities. This SAI project tackles the question of how and where to best place EV charging stations to ensure they not only meet the needs of drivers but also boost the economic resilience of small businesses and promote social equity. The project integrates theory and methods from computational social science, urban resilience, behavioral science, and complex systems to address a pressing societal need -- the equitable, resilient, and sustainable deployment of EVCSs. This project leverages large-scale datasets including mobile phone GPS, charging station usage data, and real-world intervention experiments to understand the broader social and economic impacts of EVCS placement on mobility, social dynamics, and the resilience of businesses. This complex systems approach introduces a new paradigm of infrastructure development and management that significantly extends the scope from individual behavior to social and economic community-wide effects, offering a more comprehensive understanding of the EVCS ecosystem. The optimization and visualization platform will enable agencies and businesses to evaluate hypothetical deployment scenarios, promoting a multi-dimensional approach to infrastructure design. The open-source and public-facing platform ensures that its benefits are not confined to the academic realm but are extended to diverse community stakeholders, reinforcing the project's commitment to inclusive and comprehensive urban development. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2024-09

Project Summary A robust enzyme design strategy will profoundly impact human health, such as by achieving asymmetric synthesis of pharmaceuticals not easy to be achieved by small-molecule catalysts. However, photoexcitation has rarely been considered in enzyme design. In recent years, it has been found that certain enzymes can be repurposed by photoexcitation for non-natural chemical reactions that cannot be easily achieved by small- molecule catalysts or traditionally engineered enzymes. These enzymes, termed photoenzymes, are believed to utilize photoactivatable cofactors, combined with a natural or mutated enzyme scaffold, to reach new reaction spaces. Photoenzymes have emerged as a promising new class of catalysts for non-natural reactions important in pharmaceutical synthesis, such as asymmetric radical reactions important in late-stage functionalization of drug-like molecules. However, the mechanisms of photoenzymes have not been studied well and there has not been a clear rational discovery and design strategy for photoenzymes, not only because these are emergent systems, but also because existing computational methods are not adequate. In this research program, we will develop an integrated computational framework to predict the combined effect of light, cofactor, substrate(s), and protein sequences on photoenzyme reactivity and the mechanisms that lead to this effect, and will develop a physics-informed design strategy that makes use of descriptors derived from both ground and excited electronic states to control the activity and selectivity of photoenzymatic reactions. This will fill the gap in computational enzyme design where the excited electronic states are not normally considered. In specific, we will 1) develop machine learning-enhanced simulation methods to efficiently simulate both the ground and excited electronic states of photoenzymes to assist mechanistic studies and to inform the prediction and design of photoenzymes, and 2) develop a photoenzyme design strategy centered on descriptors derived from both ground and excited electronic states computed by molecular simulations. We will use flavin-dependent “ene”- reductases (EREDs) as the prototype system for the computational tool development and testing since there have already been a collection of computational and experimental data for EREDs, where the computational data are from the PI’s group. This research program will not only deepen our understanding of photoenzyme mechanisms, but will also greatly facilitate the design and prediction of photoenzymes for non-natural reactions important in pharmaceutical synthesis. In the long term, it will also facilitate the identification of natural enzymes that may have previously unknown photo-driven reactivity, which may become new protein scaffolds for developing novel photoenzymatic reactions or become new drug targets.

NSF Awards · FY 2024 · 2024-09

Large-scale AI models that generate text and images are transforming visual synthesis and recognition tasks. These models can not only generate realistic images from texts but also serve as backbone models for object recognition, tracking, and segmentation. Despite their capabilities, they are complex, challenging to interpret, and sometimes unpredictable. This unpredictability is problematic for safety-critical applications such as autonomous driving and healthcare. Moreover, these models can easily generate copyrighted content, produce harmful content, or amplify stereotypes. Understanding such complex models, with billions of parameters trained on billions of images, remains challenging. Key questions include understanding why certain input prompts lead to failures or artifacts, how these models might amplify biases, and how the training data can impact output quality. This project seeks to provide a systematic, interpretable, rational basis for understanding and controlling the learned computations of multimodal generative models, with the potential to increase the accountable and safe use of state-of-the-art multimodal AI models and mitigate potential harms. To effectively disseminate the research, the investigators will freely release all materials (code, models, and datasets) and host tutorials, workshops, and courses to engage with the research community, enhance students’ participation at the K12, undergraduate, and graduate levels, and engage with policymakers to inform them of the latest technology and future trends. The project aims to develop a new systematic framework for visualizing, understanding, and rewriting the learned computation of multimodal generative models and leverage this knowledge to trace how the training data used influences the internal representation and, ultimately, affects the model outputs. The project will focus on three research thrusts. First, new research methodologies will be developed to visualize the internal mechanisms and the hierarchical structures of pre-trained multimodal generative models, understand their roles in different stages of the generation process, and extract visual concepts and their relationships. Second, the project will explore several model editing algorithms to manipulate these discovered concepts and relationships to pinpoint and fix inconsistencies, failures, biases, and safety concerns of existing multimodal generative models. Finally, the investigators will develop new attribution methods to assess the influence of training images on generated results based on the analysis of internal representations and show several potential applications of the attribution algorithms. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

Researchers at Northeastern University, the University of Maine, and Colby College are conducting research focused on developing and testing a technology-based solution that improves the accessibility of spreadsheet data tasks for blind and low vision (BLV) STEM employees. Data tasks include data exploration, data manipulation, and data analysis, which are often inaccessible to BLV employees due to the visual attributes of commonly used data tools and the limitations of current keyboard-based functions. Through this research, keyboard functions will be enhanced with the addition of natural language-based functions, resulting in an intuitive and accessible approach to performing complex data tasks. Proficiency in the use of data is foundational for many STEM jobs. This project addresses a longstanding accessibility issue faced by BLV employees in STEM fields Employing a user-centered approach, this project has three phases: 1) establish design considerations and guidelines for developing BLV-focused natural language (NL) interfaces that can facilitate data analytics tasks, 2) develop a prototype NL interface to support BLV users to perform data analytics tasks on STEM-specific datasets, and 3) empirically measure the practical utility of the prototype system for its ability to support BLV users in performing these data tasks and assess its acceptability and usability among target end users. With BLV STEM professionals and researchers involved across all aspects of the research, the methods include interviews, co-design workshops, Wizard of Oz prototyping, and experimental evaluation. The results of this research will be a prototype of an intelligent natural language-based system, called Auxel, that will support BLV STEM professionals with spreadsheet data tasks. This award has been made in response to the NSF solicitation “Workplace Equity for Persons with Disabilities in STEM and STEM Education” (NSF 23-593). The Directorate funds this project for Social, Behavioral and Economic Sciences’ Office of Multidisciplinary Activities and the Division of Graduate Education’s EDU CORE Research program. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2024-09

PROJECT SUMMARY (See instructions): Advances in biomedical sciences are often driven by new tools, ranging from novel experimental techniques, to innovative models, and new software. V\Jhile bibliographic data such as citation counts track the impact of research discoveries, these metrics fail to capture the full extent to which specific tools are used in science and beyond. Indeed, the impact of these tools on both the scientific community and society at large is largely invisible to traditional metrics, along with the contributions of the researchers who develop them. We need, therefore, novel databases and metrics to trace the true impact of tools, bringing them within the scope of Science of Science. Three fundamental challenges impede this goal: a data gap, i.e., a lack of a "ground truth" dataset on tools and their usage, a metrics gap, i.e., the absence of meaningful impact indicators, and a mechanistic gap, i.e., a lack of theoretical foundation to understand their adoption. Our proposal aims to tackle these challenges, by developing a data-driven approach to automatically identify the tools used in biomedicine, that we collectively call Bio Tools, and developing indicators to trace the multiple dimensions of impact they have, offering a rigorous empirical foundation for the quantitative study of Bio Tools. We start with developing a ground-truth corpus by extracting laboratory techniques, software applications, and modeling methods from existing ontologies, and identifying their mentions in the full-text of millions of documents, from publications and grants, to patents and clinical trials (Aim 1). The recovered information, along with metadata, will be assembled into the knowledge base BioToolKB, which will allow us to develop a multidimensional impact measures, and quantify the extent to which traditional metrics track or underestimate Bio Tool impact (Aim 2). Finally, we rely on these indicators and BioToolKB to understand the diffusion mechanisms of Bio Tools across biomedical fields and sectors, identifying the conditions in which some Bio Tools are widely adopted and others overlooked (Aim 3). The success of this proposal is ensured by an interdisciplinary team blending domain expertise in biomedical sciences and the Science of Science with technical expertise in bibliometrics and network science, and industry collaborators, supporting the creation of a robust technical foundation.

NIH Research Projects · FY 2025 · 2024-09

PROJECT SUMMARY The axolotl salamander is capable of perfect, scar-free regeneration throughout its body. In order to regenerate a complex biological structure such as a limb, cells at the site of injury must correctly re-establish axial patterns such that only the missing tissues regenerate. Retinoic acid (RA) is a pleiotropic morphogen that has long been studied for its roles in development and regeneration. Intracellular RA can, broadly, be secreted to neighboring cells, catabolized in the cytoplasm by cytochrome P450 family 26 (CYP26) proteins, or imported to the nucleus. In the nucleus, RA binds to its family of receptors (RARs) at retinoic acid response elements (RAREs) located in the promoters of RA-responsive genes. Upon RA binding to RARs, local chromatin remodels to modify transcription of the primary target. Inhibiting RA breakdown by CYP26 reprograms regenerating distal cells to a proximal state in a dose-dependent manner, suggesting a gradient of RA concentration along the proximodistal axis maintained by differential RA catabolism. Despite decades of study, few primary targets of RA signaling have been confirmed, and the exact genetic mechanisms by which RA establishes a proximodistal axis remain under studied. In addition, connective tissue fibroblasts have been found to be key drivers of axial re- establishment in regeneration, but their ability to store and confer positional “memory” upon injury is not well understood. Previous work has also demonstrated that fibroblasts are specifically responsive to supplemental RA. From these foundational studies, I hypothesize that differential chromatin compaction promoted by RA signaling in connective tissue fibroblasts establishes the regenerating limb’s proximodistal axis. The proposed project intends to elucidate the epigenetic regulation by RA signaling in fibroblasts and its spatiotemporal progression during regeneration through two specific aims. Aim 1 is to use single-nuclear multiomic sequencing to generate fibroblast-specific candidate genes with putative RAREs and test their responsiveness to RA. Aim 2 is to functionally test candidates through assaying the regulatory activity of RA-responsive genes and functional ablation of normal RA-driven transcriptional regulation. The proposed experiments will provide deeper insight into the molecular role of RA in connective tissue fibroblasts that generates proximodistal positional identity during axolotl limb regeneration. The findings from these experiments will enhance our understanding of the regulatory conditions behind complex tissue regeneration which promote successful regeneration in some species, such as the axolotl, but not in humans. These studies will potentially generate novel targets for regenerative therapies in humans, whose regenerative capacity is largely limited to the distal-most digit tip.

NSF Awards · FY 2024 · 2024-09

The National Science Foundation (NSF) Graduate Research Fellowship Program (GRFP) is a highly competitive, federal fellowship program. GRFP helps ensure the vitality and diversity of the scientific and engineering workforce of the United States. The program recognizes and supports outstanding graduate students who are pursuing research-based master's and doctoral degrees in science, technology, engineering, and mathematics (STEM) and in STEM education. The GRFP provides three years of financial support for the graduate education of individuals who have demonstrated their potential for significant research achievements in STEM and STEM education. This award supports the NSF Graduate Fellows pursuing graduate education at this GRFP institution. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2024-09

PROJECT SUMMARY Cardiovascular disease (CVD) is one of the leading causes of mortality in the U.S. When immigrants first arrive in the U.S., they have better CVD risk factor profiles than the U.S.-born, but within 5 to 15 years of living in the U.S., they experience a worsening of their CVD risk factor profiles. Some immigrants have lower screening, monitoring, and treatment of CVD risk factors than U.S.-born populations. Lower preventive care utilization likely contributes to their worsening cardiovascular health. Eliminating differences in CVD preventive care utilization is necessary to improve the cardiovascular health trajectory of immigrants and lessen the burden of CVD mortality in the U.S. The research objective of this career development award is to use epidemiologic, system dynamics, and community-engaged methods to understand how individual and multiple components of the environment (e.g., language barriers, provider accessibility, policies and practices, sentiment) affect immigrants’ CVD preventive care utilization. This research will also address current gaps in research by focusing on CVD preventive care utilization outcomes and using objectively measured, longitudinal electronic health record data (EHR). The research objective will be achieved through three novel aims. Aim 1 is to estimate trends and associations between state and county-level policies and practices, hate crime rates, and CVD preventive care utilization among immigrant adults using electronic health record data from a nationwide network of community health centers. Aim 2 is to develop a system map with community stakeholders that graphically depicts hypothesized causal relationships and feedback loops within the environment that affect immigrants’ CVD preventive care utilization. Aim 3 is to develop a system dynamics simulation model to estimate improvements in CVD preventive care utilization as different components of the environment are manipulated. This multidisciplinary research plan directly addresses NHLBI’s critical challenge to advance methods of assessing and characterizing exposures to understand differences in health among populations. The research aims are complemented with a training plan that builds upon the candidate’s expertise as social epidemiologist to develop the candidate’s expertise in the (1) measurement and quantification of components of the environment, (2) analysis of longitudinal EHR data, (3) use of system dynamics simulation methods, and (4) use of community-engaged research and community-based system dynamics methods. Together, the research and training plans will prepare the candidate to become an interdisciplinary, independent research investigator whose career is aimed at improving the cardiovascular health trajectories of immigrants in the U.S.

NSF Awards · FY 2024 · 2024-09

This project has the potential to enhance the impact of NSF sponsored research by catalyzing the formation and success of small business concerns founded by the NSF Small Business Innovation Research and Small Business Technology Transfer (SBIR and STTR) programs. The companies formed by these grantees are working to solve critical challenges facing our society, namely the generation and storage of renewable energy; the development of new semiconductor, communications, computing and quantum technologies; the advancement of national security technologies; and innovations in biological and medical technologies. By facilitating international cooperation on deep-tech startup creation, this project will increase the economic competitiveness of the U.S. and encourage the development of enhanced infrastructure for research and education. Finally, this project will conduct activities to encourage and support participants reflecting the diversity of the United States in founding and leading deep-tech startups. This project represents a tremendous opportunity for U.S. deep-tech startups to advance through international partnerships. While opportunity exists there has been very little work to date to identify specific methodologies that could help these startups to find, develop, and execute such partnerships. There is also a lack of information about potential opportunities that are currently available or that could be created in various partner countries. This project will address these knowledge gaps and test potential solutions with U.S. startup founders and prospective partners. Successful completion of this project will advance knowledge about the needs of NSF SBIR/STTR grantees and other deep-tech startups and on methods to leverage international collaboration to accelerate the formation and growth of these companies. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

Part of being a good scientist is not just discovering new things but also communicating those findings to others. Scientists are taught how to navigate ethical issues around consent, honesty, and transparency in the conduct of research. The investigators of this project aim to expand this educational framework to also include teaching of how to be ethical statistical communicators, especially for the charts and graphs that are often the first or only contact that audiences have with scientific data. The wrong graph can hide important information, mislead its audience, or even dehumanize its subjects. Therefore, this project seeks to create a guide, for use by scientists from all STEM disciplines, for ethical statistical graphics: to create charts that are not just effective at showing data, but that transparently, honestly, robustly, and, above all, mindfully communicate information. The project also provides opportunities for graduate students to participate in the development and testing of guidelines for ethical and effective practices for statistical graphics. This project focuses on the development of a guide for ethical and effective practices for statistical graphics. The guide will be formed through a combination of interviews with existing practitioners and ethicists, and it will be validated through a set of laboratory studies. The investigators will structure the ethical principles that will make up the core of the guide through the lens of virtue ethics: ways that scientists can embody a set of proposed virtues in their work. The investigators intend on both creating guidelines based on a survey of prior work but also conducting their own crowdsourced experiments to empirically validate these guidelines, especially in places in which there are conflicts in values or virtues. For instance, the increased transparency that comes from including detailed statistical uncertainty information incurs a cost in complexity, and it limits the intended audience to those that understand how this uncertainty information was modeled: how do varying levels of uncertainty impact perceptions of scientific work, and is there a “sweet spot” between including too much uncertainty that can overwhelm the intended viewer, and too little that can mislead viewers or sweep potential issues around the strength or reliability of scientific findings under the rug? Once the investigators’ guidelines have been collected and the strength of the evidence assessed, the investigators will deploy the resulting guide for ethical statistical graphics in classrooms and workshops as well as online for use in training the next generation of scientists. It will also be used to inspire existing scientists and science communicators to think more deeply about the ethical issues underlying their work. This project is jointly funded through the ER2 program by the Directorate for Biological Sciences and the Directorate for Social, Behavioral and Economic Sciences. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

The Clostridia are a diverse group of bacteria that range from agents of human disease to industrial microbes used for renewable production of biofuels and biochemicals. Despite the importance of these bacteria, the majority cannot be genetically engineered, and even those that can are notoriously difficult to work with. This project develops fundamentally new approaches for genetically manipulating these bacteria, allowing scientists to study and control their unique physiology and metabolism at an accelerated pace, with far reaching impacts on sustainable biomanufacturing and human health. The research will provide interdisciplinary training for graduate students and a postdoctoral fellow. Additionally, aspects of the project will be incorporated into educational materials for K-12 and undergraduate students to expose them to career options in STEM fields. The tools developed will be disseminated to the broader community through conference workshops, online resources, and a new hands-on summer course. This project leverages state-of-the-art tools including CRISPR, phage recombinases, transposons, and directed evolution to develop new approaches to overcome the two main limitations in Clostridial engineering: i) low plasmid transformation efficiency; and ii) the reliance on plasmid-based homologous recombination for genome modification. The central premise of our team’s approach is that bringing together multiple labs with expertise in diverse Clostridia during the development and validation phases will result in more robust and portable functional genomics tools, maximizing their utility for the broader community. To ensure the broad dissemination of these tools and to democratize Clostridial genetics, we will create and maintain ClostridiaWiki, a community resource that will house detailed protocols and videos, as well as provide a forum for technical discussion between researchers. We will also run workshops at key conferences attended by Clostridia researchers. A new two week-long intensive Clostridial Genetics Short Courses at a state-of-the-art training lab at Northeastern will provide trainees and early career faculty lab-based instruction in Clostridial genetics, as well as seminars from leading local Clostridial researchers and professional development activities. Finally, in collaboration with the Northeastern and Tulane Centers for STEM Education, and the Tufts Pathway-to-the-PhD program, we will incorporate scientific aspects of this project into outreach activities aimed at broadening participation of URM K-12 and college students in STEM fields. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

The dynamics of human behavior play a crucial role in the spread of epidemics. While much research has focused on individual reactions to risks and policies, this project examines how groups of people, such as households, communities, or organizations, demonstrate coordinated risk-mitigating behavior and make collective decisions during an epidemic. These group-level behaviors can significantly impact the trajectory of an epidemic, beyond what can be captured by aggregating individual behaviors. By studying group behaviors, such as the formation of social bubbles and changes in risk-mitigating norms and conventions, this research aims to create better mathematical models that reflect real-world social interactions. These models will help scientists and policymakers develop more effective strategies for managing epidemics, ultimately saving lives and reducing social and economic impacts. Additionally, insights from this research could inform policies on a range of issues including gun violence, opioid abuse, disaster response, and community resilience, where group behaviors play a critical role. The research concentrates on two main questions: 1) How can mathematical models and scalable computational algorithms be created to incorporate group-level behavioral responses in epidemic models? 2) How much do group-level responses significantly influence pandemic trajectories, and what are the resulting policy implications? The team plans to jointly work on several interconnected research thrusts. They will build mathematical foundations using a three-level network model and cooperative game theory to incorporate group-level behavioral responses, such as the formation and transformation of pandemic social bubbles and localized risk-mitigating norms within pandemic models. Next, they will create computational models that enable scalable and interpretable execution of these network-based approaches, developing dynamic networks using geospatial data and designing network downscaling algorithms to improve simulation efficiency. The team will use causal identification based on various natural experiments to estimate the input parameters of the models, focusing on empirically measuring perceived risk, peer effects on interaction networks, and the formation of social bubbles. Finally, they will implement and validate the model comprehensively at the county level in the US and at a more granular level in Boston neighborhoods, examining the policy implications of group-level behavioral responses. This award is co-funded by DMS (Division of Mathematical Sciences), SBE/SES (Directorate of Social, Behavioral and Economic Sciences, Division of Social and Economic Sciences), and SBE/BCS (Directorate of Social, Behavioral and Economic Sciences, Division of Behavioral and Cognitive Sciences). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2024-09

Lipids in drug delivery systems and food can have a tremendous impact on the systemic absorption of orally delivered compounds, including drugs, nutrients, and toxins. However, this impact cannot be quantitatively predicted a priori, and underlying mechanisms are poorly understood. The ability to quantitatively predict the impact of lipids in drug delivery systems and food on oral compound absorption could revolutionize the oral formulation development process, significantly streamlining resource intensive drug development, enabling design of effective nutritive supplements, and increasing our understanding of how food composition and structure impact health in an era of rapidly evolving food availability. The Carrier lab is focused on applying engineering analysis to intestinal systems for enabling effective oral delivery, as well as understanding the impact of ingested materials on human health. To date, the lab has developed a mechanism-based modeling framework to predict the impact of lipids on oral compound absorption. Major goals of this project include: 1. Determining whether systems-based modeling approaches, capturing kinetics of key processes occurring in the gastrointestinal tract, can be used to describe and predict the impact of lipids in actual oral delivery systems and food with complex composition and structure on absorption of orally delivered compounds. 2. Revealing mechanisms by which lipids impact co-transport of orally delivered compounds across the intestinal mucosa into lymph or portal circulation, together with appropriate mathematical expressions to describe these processes. In addition to the laboratory`s track record of developing and translating mechanistic, predictive models of oral delivery and existing preliminary modeling framework, the lab is extremely well-positioned to markedly advance our understanding of oral absorption processes, given its expertise and experience in design of advanced engineered experimental intestinal models and mechanisms for exploring the impact of the mucosal barrier on oral drug delivery, both of which will be leveraged in this project period. In addition, the PI has an extensive track record of multidisciplinary collaboration, including working with multiple industrial partners, to best ensure that research outcomes translate to practice and are impactful to broad patient populations.

NSF Awards · FY 2024 · 2024-09

The broader impact of this Broadening Participation for Engineering Track-2 (BPE-Track-2) proposal addresses the critical need to enhance AI competency among K-12 educators in STEM fields. As AI technologies rapidly advance, there is a growing demand to prepare students for an AI-driven future. The project is to transform K-12 education by equipping teachers with the knowledge and skills to incorporate AI into their classrooms. The project ensures that students, particularly those underrepresented in STEM, are prepared for an AI-driven future. By providing comprehensive professional development (PD) for teachers, this project fosters a learning community that supports the development and implementation of AI-effective teaching practices. The project enhances classroom teaching and promotes equity in STEM education. It also contributes to a more technologically proficient, and AI-informed STEM workforce. The project aligns with the National Science Foundation's mission to promote scientific progress by addressing the growing demand for AI literacy in education. Through the development and dissemination of AI best teaching practices and an innovative AI-focused teacher PD model, this project has the potential to transform high school education on a national scale, aligning with NSF's mission to advance the nation's prosperity and maintains its competitive edge in the global economy. The proposed project will establish a three-year PD program for high school teachers focused on Generative AI in education. Each year, the program will recruit 10 teachers, prioritizing those serving students underrepresented in STEM fields. The project begins with a pre-survey to assess current AI knowledge, followed by introductory modules on AI history. Teachers will then participate in an intensive one-week summer PD session on AI, which will be supplemented by year-long support throughout the academic year, including monthly collaborative meetings and ongoing content delivery. Teachers will collaborate with the project team to develop and implement AI-informed lesson plans in their STEM classrooms. The program employs a hybrid approach, combining direct instruction with the fostering of a learning community across schools and districts. Key activities include monthly collaborative meetings, callback sessions for feedback, and the creation of online tools for nationwide dissemination. The project's outcomes include a scalable teachers PD model, the creation and implementation of AI-informed lesson plans for use in high school STEM classrooms, and broad dissemination of resources through online platforms. This comprehensive approach aims to cultivate AI competency in education, ultimately preparing the next generation for careers in an AI-driven world. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

This project aims to create micro-mirrors for applications in endoscopes that enable controlling the depth-of-focus and imaging in real-time. These instruments can guide surgeries to distinguish normal and malignant tissues with sub-cellular resolution. State-of-the-art endomicroscopy utilizes MEMS scanning mirrors for single-axis and dual-axis confocal microscopy. However, the depth of focus and scanning speed are severely limited because of the actuation mechanism. The most common actuators have been based on conventional electrostatic mechanisms that face two bottlenecks: small range of motion and limited speed. The team of investigators will overcome those limitations by using electrostatic levitation that enables large strokes away from the substrate. To enable high-speed scanning, investigators will adopt ferroelectric material to control spring stiffnesses. This property enables achieving a wide range of scanning speeds at any elevation. The large range of motion because of electrostatic levitation, and stiffness tunability using ferroelectric material will permit the development of tunable MEMS scanners that can revolutionize endomicroscopy for real-time in vivo imaging and 3D depth sensing. The new microscanner increases the depth of focus and enables deep tissue penetration over a large FOV with sub-cellular resolution. To educate a wide range of learners, the investigators develop workshops for demonstrating the basics of micromirrors and present them to elementary schools as well as undergraduate students. Investigators will involve undergraduate students from a broad group of students. This project will create new knowledge on the interaction of electrostatic levitation with ferroelectric polarization switching. Based on this new knowledge, investigators will create tunable high-speed and large-stroke MEMS mirrors. For more than forty years, MEMS mirrors for imaging applications have been based on conventional gap-closing mechanisms that severely suffer from a limited range of motion and pull-in instability. To address those issues, a team of researchers will introduce electrostatic levitation electrodes that allow the actuator to move away from the substrate and have its motion become a linear function of applied voltage. To achieve tunable high-speed actuation, the team will incorporate springs made of integrated ferroelectric material that enables stiffness tuning to trigger modes of interest, e.g., titling or out-of-plane at desired frequencies. The merger of electrostatic levitation and ferroelectric polarization switching creates challenging behaviors that have prevented researchers from adopting it for micro-mirror applications. Investigators will present a computational, analytical, and experimental platform to present a fundamental understating of the underlying multiphysics of the system. In addition, the team will create a MEMS mirror that achieves high-speed large strokes and rotations from four actuators on its periphery. The micromirror prototype will pave the way for its future application in real-time in vivo microendoscopy. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

Genomes fold into architectures that are characteristic of both cell phase and cell type. The spatial organization of genes orchestrates interactions among genetic regulatory elements that ultimately contribute to gene regulation in organisms and tissues. Understanding the principles underlying genomic architecture and its effect on transcriptional regulation is a major undertaking with far-reaching implications in basic science, medicine, and technology. As experimental information about the spatial conformation of genes becomes available, accurate theoretical models are increasingly needed to interpret this data. Unfortunately, theoretical models for genes are being outpaced by developments in experimental approaches. This is partially because genes are molecular systems that are simultaneously too big to be studied with conventional molecular dynamics simulations and too heterogeneous to be tackled with polymer physics-based approaches. This goal of this award is to fill this gap by developing a new theoretical approach to studying genes using the framework of Associative Memory Hamiltonians. The proposed theoretical model would enable molecular dynamics simulations of tens of thousands of DNA base pairs together with the proteins decorating the fiber, representing specific genes or gene clusters. The model allows integration of a vast amount of information that is already known for the smaller molecular scale to study the mostly unknown conformations of genes, much larger systems. The objectives of the project are: (i) to faithfully model the mechanics of chromatin at gene scale, given information about the positioning of proteins that bind along the DNA polymer, (ii) to study how genomic conformational ensembles are affected by supercoiling, nucleosome occupancy, specific histone variants, and the presence of other proteins such as polymerases or transcription factors, (iii) to elucidate how essential physical contacts related to the operation of the genome, e.g. enhancer-promoter interactions, are orchestrated by the three-dimensional architecture of genes. In addition, the PI participates to several existing outreach programs aimed to fostering participation of underrepresented groups to research in biophysics. Undergraduate students recruited through these programs will participate in the research proposed with projects involving developing scientific software. The software engineering skills acquired through these projects will increase students’ competitiveness in the pursuit of graduate education and employment. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

This award supports research that enables a new dynamic model for fluid-structure interaction (FSI) systems with a focus on large wind turbine blades, thereby promoting the progress of science, and advancing prosperity and welfare. The project will promote safe design of next-generation offshore wind turbine structures by enabling slender and lighter blade designs. The research will provide indications about various wind turbine blade aeroelastic instability thresholds along with the most appropriate simulation and analysis approaches for novel designs of longer and therefore more efficient wind turbine blades. Although flow-induced instabilities have been predicted to occur for such new wind turbine blade designs, predictions are often based on deterministic models without the influence of flow turbulence and load variability. This project will address this critical gap by combining experimental measurements and theoretical modeling to derive a novel model that accounts for the influence of turbulence on the onset of instability and post-critical behaviors. This research is timely since the current energy plan of the United States strongly emphasizes the need for alternative and sustainable energy production by offshore wind energy. Integrated research and educational initiatives will complement the activities. The findings of this research will be disseminated, at different levels, by integrating the research into the outreach programs for K-12 students and teachers, creating new modules for different courses, hosting high school classes, and broadening research opportunities for students from under-represented minority groups. This research aims to make fundamental contributions to accelerate the use of stochastic and probabilistic structural dynamics to examine pre- and post-critical behavior of fully-coupled FSI systems with asymmetric structures and subjected to three-dimensional flows that can undergo nonlinear dynamic instabilities. It will achieve this goal by producing models that describe turbulence effects and more accurately considering the aeroelastic loads, which are relevant to highly flexible and asymmetric structures, such as a wind turbine blade. The existing aeroelastic load models are mainly for basic-shape airfoils (symmetric, small thickness idealized surfaces), and their flow parameters are based on experiments conducted on airfoil cross sections that represent long structures, under two-dimensional flow conditions. The researched modeling will include the asymmetries, twists and variable thicknesses that are typical of modern wind turbine blades but also applicable to a wider range of structures similar to wind turbine blades. These asymmetries and twists result in a highly three-dimensional turbulent flow. Several series of experiments will be conducted both at small and large scales. Information on flow forces and turbulence intensities for each case will be collected to inform a semi-empirical stochastic model for the onset and post-critical instability of the structure. The model will account for three-dimensional “rotationally sampled” flow features and will be validated at two separate experimental scales. These two sets of aeroelastic experiments will enable model verification for larger and full-scale structures. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

Education researchers have access to more extensive and heterogeneous data sources for their research and assessments, which requires skills in advanced cyber-infrastructures. Artificial intelligence (AI) can help improve the quality of educational research and assessment. This kind of research and assessment is invaluable in advancing national interest by enhancing the ability to answer research questions such as the effectiveness of education policies and pedagogy techniques and closing the achievement gaps. Utilizing AI in education requires additional skills beyond conventional statistics training education researchers, school administrators, and policymakers receive. This project addresses the fundamental issues of training users to use advanced cyber-infrastructure, such as cloud computing systems, to deal with the challenges of working with large quantities of education data. The training materials, software tools, and hands-on project assessments developed as part of this project help prepare future educational researchers in learning analytics to use advanced cyberinfrastructure systems in the cloud. The other potential benefits include expanding the utilization of cyberinfrastructure resources beyond the traditional natural science researchers to involve other social science researchers in education to serve national needs. This project, AI4EDU, aims to develop innovative training materials for education researchers to enable them to utilize AI in educational research and assessment using cloud infrastructures. AI4EDU consists of three integrated thrusts to address this challenge. The first thrust is the development of educational materials that introduce critical aspects of planning, configuring, and utilizing cloud computing resources and frameworks (e.g., Hadoop, federated learning) to support various educational analytical tasks. The second thrust is to develop tools in data quality, cloud monitoring, cloud planning, and configuration to support utilizing cloud services. The last thrust is to design sample projects with accompanying datasets for real-world, hands-on training. In addition, AI4EDU includes a public repository to collect and share machine learning programs and datasets tailored for various educational research tasks to help build up the community of users. The AI4EDU project helps support the AI for Education initiatives by bridging the gap between the analytical techniques taught in the classroom and the tools and skillsets needed to work with data in education. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

About 15-20% of the adult population identifies as neurodivergent. These individuals offer immense unrealized potential as employees; however, they are a vulnerable community subject to extreme social and systemic inequities in jobs and higher education. Neurodivergent adults experience chronic unemployment and underemployment. When employed, they are underrepresented in management and leadership roles within organizations. Solving this complex problem goes far beyond the reach of any single discipline, but requires theories, methodologies, and approaches that encompass policy, organizational, group, individual and technological insights as well as meaningful involvement of neurodivergent individuals. The objective of this planning proposal is to assemble a team of researchers, employers, educators, advocates, and neurodiverse individuals to study, develop and disseminate organizational and technological evidence-based practices to better support the advancement of neurodivergent individuals in meaningful, productive work and increase worker productivity, job satisfaction, and career advancement. This project brings together an interdisciplinary team of researchers with expertise in artificial intelligence, behavioral science, data science, game design, organizational psychology, physical therapy, rehabilitation science and special education, to collaborate with advocates, educators, employers, and neurodivergent individuals to transform the current state of employment for adults who identify as neurodivergent. Building on previous NSF funded research, the work described in this planning proposal will create a muti-university, multidisciplinary Center for Neurodiversity Development and Advancement that includes both researchers and key stakeholders collaboratively designing research questions and developing solutions. Integrating scientific knowledge from educational, organizational, technological, and psychological research, each participating university capitalizes on its unique strengths and builds a collaborative team with neurodivergent individuals and advocates included as partners. Products of the center will include research to solidify factors underlying lack of employment opportunities, development of supports to enhance access to higher education and job training in collaboration with the neurodivergent community, development of strategies to facilitate meaningful employment and career advancement, and education to organizations and other key stakeholders within the broader community to promote employment equity. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

In the data-driven world that we live in, sharing digital information is a key component underpinning a vast body of technologies. Low-latency, high fidelity access to information is central to algorithms that impact how we work, are entertained, how we travel, and our healthcare. Systems that, in particular, rely on wireless communication to deliver their services have become ubiquitous. With the increase in data transmitted over the air, however, the central resource that they depend on, spectrum, can become congested with multiple communications overlapping and negatively impacting each other. This project brings together diverse researchers from Northeastern University (NU), the Massachusetts Institute of Technology (MIT) and Boston University (BU) to develop methods that improve communication performance in shared, congested, and contested spectrum bands. The presence of interference in communications detrimentally impacts the throughput and reliability of systems. Interference and noise are often used interchangeably as they are commonly lumped together as general deleterious effects that corrupt communications. Interference, however, has a more structured form than noise. Central to this project is developing new means to leverage that structure to improve communication systems. By enabling more efficient use of scarce resources, more services can reliably co-exist, advancing national health, prosperity and welfare. By developing techniques that are receiver-only, it allows both backward compatibility and graceful adoption paths. Interference management motivates substantial engineering effort at all levels, from hardware design, to signal processing, to error correction, to retransmission, and resource allocation protocols. A traditional approach to managing interference is to consider its impact as being part of noise. This project aims to do more, leveraging the structure of interference to improve performance through receiver-side approaches only, thus circumventing barriers to technological adoption. When a modulated communications signal experiences interference that arises from other modulated communications, those characteristics can be taken into account. Even when an interferers' modulation may not be discerned, the interference can influence the noise experienced by a receiver in semi-predictable ways that can be exploited by a receiver. When interference is due to the presence of other communication systems where individual interferers' modulation can be detected but the signal not decoded, unlike in a multiple user system, this project proposes an approach that takes into account both noise and the restricted forms the interference can take. When channel and modulation may not be available at the receiver, interference will still have characteristics that are different from, e.g., Gaussian noise. The statistical characteristics of such interference can be used to improve forward error correction decoding, enabling reliable communication with less overhead, which this project explores. When interference is due to signals that vary more slowly than the communication, such as from electronic devices, the receiver cannot rely on knowledge of the structure of the interference, other than the fact that it will exhibit a slowly varying profile. In that case, this project aims to discover post-decoding the interference experienced by some signals and use it as a starting point to remove pre-emptively at least partially that interference from other signals that are proximate in time, and thus subject to a similar interference. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

To address the critical need for a more diverse and inclusive engineering workforce, this project will establish a pioneering university-industry-student partnership aimed at facilitating equitable access and transition into civil engineering careers for individuals with disabilities. Despite calls from the National Science Foundation and the National Institutes of Health, people with disabilities remain severely underrepresented in STEM fields. In industry, engineers with disabilities constitute less than 10 percent of the workforce and are less likely to be employed than non-disabled engineers; those who are employed generally experience lower pay. To-date, scholarship examining the accessibility of academic institutions has focused on the programmatic experiences of undergraduate engineering students with disabilities, with little to no work continuing past the point of graduation. As a result, this project, aligned with the National Science Foundation's commitment to fostering inclusivity and innovation in engineering education, represents a pivotal step towards broadening the participation of engineers with disabilities in the civil engineering industry. Focused within the civil engineering sector, pivotal to national infrastructure development, this endeavor will lay the groundwork for transformative programming supporting disabled students' transition from academia to professional practice. People with disabilities have been referred to as “the original lifehackers” due to the innovative ways they alter everyday products, systems, and spaces to access a world not built for them. While innovation and problem solving are core competencies in engineering, the role of people with disabilities as engineers has not been realized for many reasons. These reasons include social and professional stigma and a lack of support structures that facilitate the entry of engineering graduates with disabilities into the workforce. Beyond diversification, the project aspires to promote genuine inclusion, illuminating the underrepresented cohort of disabled engineering students and laying foundational steps for accessible engineering education and practice. This planning grant will contribute to a deeper understanding of existing scholarship and current industry perspectives, provide a framework for developing partnerships between academia and industry, and blaze a trail forward for creating a more diverse and inclusive engineering workforce through the following outcomes: (1) synthesizing relevant literature; (2) identifying and engaging industry stakeholders to explore collaborative tensions and synergies among industry stakeholders; and (3) developing a robust research agenda for the next phases of the project. In Phase 1, we will employ systematic review techniques to conduct a literature review to examine the research landscape of the engineering school-to-work transition, industry practices for hiring people with disabilities, and university/industry partnerships. In Phase 2, we will conduct interviews to help us foster interpersonal relationships with the industry partners recruited in Phase 1. In Phase 3, we will apply the outcomes identified in Phases 1 and 2 to establish a robust research agenda for project continuation. By bridging academia and industry, this research will enrich scholarship, provide a framework for sustainable partnerships, and foster a more inclusive engineering workforce. Moreover, this initiative holds broader impacts by pioneering inclusive career pathways that destigmatize disability in industry, promoting transparency, and emphasizing the unique contributions of individuals with disabilities in infrastructure design. Most importantly, it will provide the critical first steps to creating inclusive and accessible pathways to and through engineering for all engineering students. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2024 · 2024-09

This Smart & Connected Communities (SCC) project supports research to address the crucial challenge of urban mobility for aging population by leveraging artificial intelligence (AI) and virtual reality (VR) technologies. Enabled by realistic urban simulations, this project aims to improve how cities accommodate the mobility needs of older adults, making urban environments more accessible and inclusive. The novelty lies in the methodology for transforming infrastructure planning, design, and operation through advanced technologies while emphasizing social equity and user experience. The project outcomes could be used to foster inclusivity in civil infrastructure systems, enhance quality of life for the elderly, and provide educational opportunities to inspire the next generation of engineers and scientists. This research tackles an often-overlooked problem in many cities that older adults face when navigating complex urban spaces. It becomes increasingly critical as the global population ages and civil infrastructures remain underfunded. The project employs a novel data-driven framework that integrates temporal point process-based deep learning (TPP-DL) with VR to self-generate dynamic, immersive simulations. These tools not only reflect the actual mobility challenges experienced by older adults but also allow for identification and mitigation of biases in infrastructure planning, design, and operation. By incorporating community feedback and utilizing edge computing for real-time data processing, the project ensures that the solutions being developed are both effective and practical. The long-term goal is to create smarter, more inclusive cities. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2024 · 2024-08

The Engineering for Women’s Health Conference addresses historical disparities in women’s health research by establishing a dedicated platform for interdisciplinary collaboration and discourse. Despite the historical emergence of gynecology, progress in women-specific health issues has experienced a notable lag. This conference, situated in Greater Boston, a nexus of cutting-edge institutions, emphasizes the crucial role of engineering in advancing women’s health. Specific Aim 1 focuses on providing a platform for prominent engineering researchers and experts in women’s health in the Greater Boston area to present and discuss their latest findings. This initiative aims to not only foster collaboration and identify gaps in scientific knowledge but also underscore the pivotal contributions of engineering to innovative solutions. Specific Aim 2 aims to engage the broader community, particularly members in training, in a dialogue concerning the latest advancements in women’s health research. Through traditional presentations, networking events, and the creation of a publicly accessible “Conference Brief,” the project will extend the impact of collective efforts to a broader audience, highlighting the importance of incorporating engineering perspectives in addressing women’s health challenges. Aligned with the mission of the National Institute of Child Health and Human Development (NICHD), this conference directly contributes to the improvement of the health and well-being of women, underscoring the unique role that engineering can play in enhancing health outcomes for women and children. Led by Principal Investigator (PI) Amini and faculty co-advisor Bellini, the project involves a collaborative effort with a dedicated team of PhD students. Distinguished speakers selected from the Greater Boston area will significantly contribute to the overarching objectives of the conference. In essence, through the Engineering for Women’s Health Conference the PI and the organizing team endeavor to bridge historical gaps, push the boundaries and inspire innovation within the context of engineering, and contribute to the advancement of health outcomes for women.

NSF Awards · FY 2024 · 2024-08

Artificial intelligence (AI) algorithms underpin social media. Algorithms sift through a large inventory of content, deciding what appears at the top of each user's feed. These social media AI systems can shape people's beliefs, affect their well-being, and change their behaviors. These consequences accrue to individuals, but also aggregate at the societal level where the value of social media AI has been stubbornly difficult to square with the societal harms that they produce. Such issues are in part due to the individualist values embedded in how social media AI software operates, maximizing each user's individual experience–-as inferred, for example, through their likes, retweets, and surveys–-at the cost of societal preferences, such as community health and civic engagement. This project aims to shape an alternative future where social media AI software aids us in achieving societal goals, by demonstrating the feasibility of integrating such societal objectives into social media algorithms used to prioritize content in users’ feeds. The project goal is to create a method that can build translational science on top of social science and computer science, and develop engineering solutions that can be deployed at scale on social media, if desired. This project will develop techniques for encoding societal values into social media ranking algorithms. Our multi-disciplinary team of researchers aims to 1) introduce a novel method that leverages the precise language of definitions and measurements of the social science constructs to build algorithmic objective functions using large language models (LLMs), referred to as societal objective functions, which can be deployed broadly as weights in social media ranking algorithms; 2) create a pluralistic algorithmic library of such societal objective functions based on rigorous and empirically validated social science theory articulating a broad space of values; and 3) build methods to integrate multiple potentially-competing values and understand the trade-offs between them. To achieve these goals, the project will weave together social science and computer science insights. Social science research will articulate the design space of societal values at play, as well as careful definitions and measurements of each of these values. Computer science research will translate these social scientific insights into AI models that agree with community ratings on the values expressed in social media content, enabling integration into feed ranking algorithms. By conducting large-scale field experiments with diverse populations, this project will provide empirical evidence on the impact of integrating a pluralistic library of societal values into such algorithms. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.