James C. Lester - US grants

The university is developing a course on the principles and technology of multimedia, intended for juniors and seniors in Computer Science and Computer Engineering. This course has a significant lab component to demonstrate major ideas, generate enthusiasm, and develop hands-on skills. Major subjects include signal processing, compression, scheduling, data retrieval, interface design, device interfacing, networks, media (sound, music, voice, images, graphics, video), and applications programming. Equipment for this lab consists of networked personal computers, programming and applications software, and I/O devices. The course's progress is being measured with the help of a group of outside experts. The materials developed in this lab include lab exercises, training materials, recommendations on hardware and software, and software for functions that are not commercially available. These materials will be made available electronically to other teachers.

Knowledge-based learning environments can provide highly customized problem-solving experiences that are tailored to the individual needs of each student. This project creates the technology for a new generation of knowledge-based learning environments by developing real-time multimedia explanation generators with animated pedagogical agents. Prominently featured in learning environments, animated pedagogical agents can observe students' progress, provide them with visually contextualized problem-solving advice, and play a powerful motivational role. The project has three major thrusts: (1) developing a computational model of real-time multimedia explanation planning for dynamically creating customized multimedia explanations; (2) developing a computational model of animated pedagogical agents; and (3) conducting empirical evaluations of the pedagogical effectiveness of these models. In addition, the project also develops a teaching laboratory for multimedia technology and initiates a new multidisciplinary course on Knowledge-Based Multimedia Learning Environments. This work significantly advances the state-of-the-art in knowledge- based learning environments and human-computer interaction. Coupled with rapid advances in affordable multimedia hardware, the successful deployment of these technologies will have a deep and lasting impact on students at all levels. By achieving significant gains in learning effectiveness, these technologies will bring about fundamental improvements in both the classroom and the workplace on a broad scale.

9720395 Lester, James C. North Carolina State University Learning and Intelligent Systems: Animated Pedagogical Agent for Constructivist Learning Environments Constructivist learning environments facilitate discovery-based learning through creative problem-solving experiences. To do so, they should provide scaffolding in the form of highly contextualized problem-solving advice that is customized to each learner. Perhaps the most intriguing vehicle for providing such dynamically individualized scaffolding is the technology of animated pedagogical agents. Featured prominently in learning environments, they couple key feedback functionalities with a strong, lifelike visual presence. The objectives of the proposed research are two-fold:(1)Creating a new generation of intelligent constructivist learning environments with animated pedagogical agents; and(2)Providing a comprehensive, data-rich account of the cognitive processes and results of interacting with constructivist learning environments with animated pedagogical agents. Employing a broad array of quantitative and qualitative measure of learning, the large multidisciplinary research team is working closely with students and teachers in a local school system to conduct extensive empirical studies of learners' interactions with these environments. This work makes significant contributions to the growing body of research that blends technology,cognitive science,and learning and teaching. First,it creates a rich constructivist frame work for learning-by-designing that couples exploratory learning with strong scaffolded support. Second, by focusing on analyses of the cognitive changes that learners undergo,it identifies precisely which technologies and conditions contribute most to effective learning.Third, because it employs a participatory design process that facilitates the technology's incorporation into school settings in authentic ways,it is expected that it will have a direct and significant impact on classroom experience well into the 21st century.

This is an award to study pedagogical agents. Pedagogical agents are embodied software agents that have emerged as a promising vehicle for promoting effective learning. They provide customized problem-solving experiences and advice that are precisely tailored to individual learners in specific contexts. By co-habiting a rich inquiry-based learning environment with learners, pedagogical agents can observe learners' problem solving activities, offer situated advice, and actively support learners' iterating through cycles of questioning, hypothesis generation, data collection, and hypothesis testing. However, inquiry-based learning also presents a significant challenge: the very "openness" of the learning environment introduces multiple sources of complexity into tutorial planning. To address the complexities associated with supporting inquiry-based learning, this project proposes the use of Bayesian pedagogical agents that leverage recent advances in Bayesian and decision-theoretic computational models of reasoning to promote self-regulated learning experiences that are both effective and engaging. It will develop a full suite of Bayesian pedagogical agent technologies for inquiry-based science learning environments. To promote effective and engaging learning processes and outcomes, it will create Bayesian pedagogical agents that leverage probabilistic computational models that systematically reason about the multitude of factors that bear on decision making to infer learners' beliefs, goals, and plans, including strategy use, from their problem-solving actions. By introducing pedagogical agents into the visually engaging environments that typify high-end game platforms and embedding them in dynamically generated science narratives, it will address the complementary goals of achievement and engagement. The project will also provide a comprehensive account of the cognitive processes and results of interacting with Bayesian pedagogical agents in inquiry-based science learning by conducting extensive empirical studies. To understand the cognitive mechanisms by which self-regulated inquiry-based science learning occurs with middle school students interacting with Bayesian pedagogical agents, the project will take a multi-method approach to investigating the use and effectiveness of Bayesian pedagogical agents. In both controlled laboratory and classroom-based field settings, these studies will investigate the central issues of self-regulation with respect to both achievement (science content knowledge, transfer, and effective strategy use, including strategy selection and strategy shifting) and engagement (self-efficacy, situational interest, and mastery orientation with an emphasis on persistence) to determine precisely which technologies and conditions contribute most effectively to learning processes and outcomes. The intellectual merit lies in the marriage of Bayesian networks and education research. The broader impact is in the promise of vastly improved technology for education.

The project builds on the earlier developed Crystal Island and draws upon intelligent tutoring and narrative-centered learning technologies to produce a suite of intelligent game-based learning environments for upper elementary school science students. The games explicitly model student knowledge and problem solving and dynamically customize feedback, advice, and explanation as appropriate. Unlike its predecessor, the platform is multi-user so it can support collaboration; offer dynamically generated feedback, advice, and explanation; and provide a pedagogical dashboard that generates student progress reports.

Students navigate rich storyworlds setup by engaging narratives, interact with a large cast of characters in the game, and manipulate artifacts in the environment in the course of solving problems. They are entering an intelligent, game-based learning environment that is a laboratory where researchers can investigating various approaches through which complex problem-solving skills can be most effectively acquired. Because such environments include multiplayer interaction and voice communication, research can study complex communication in the context of collaborative problem solving. With multiple students coordinating their efforts to solve problems, intelligent game-based learning environments create situations that require effective collaboration skills. Through these, engagement is also studied, which includes motivation, situational interest, presence, flow, goal-orientation, and self-efficacy.

Representation is a central construct in the creative process. Exploring alternate representations can significantly increase the quality of artifacts created, and using multiple representations enables creators to broadly consider the multifaceted nature of the problem spaces they explore. Narrative offers an ideal laboratory for investigating multiple representations in the creative process because stories can be expressed in rich text (a static, uni-modal representation) re-represented in animated stories with accompanying narration and spoken dialogue (a dynamic, multimodal representation). The objective of the proposed research is to design, build, and empirically evaluate an interactive creativity environment that facilitates the exploration of alternate representations in the creative process. In particular, the proposed work will focus on the Narrative Theatre, an interactive narrative-centered creativity environment.

The project will yield a cognitive account of creativity that will inform the design of next-generation creativity environments. By promoting rich interactions that are simultaneously effective and engaging, it will find broad application in education and training technologies. The project will increase the participation of women and underrepresented minorities. It will involve a diverse population in the Narrative Theatre user studies, it will involve women in all aspects of the research, it will train undergraduates through involvement in formal and informal research exposure efforts, and it will increase the participation of women and underrepresented groups in computer science through interaction with the STARS Alliance.

Because of the growing recognition of the role that emotion and motivation play in learning, affective computing has become the subject of increasing attention in research on interactive learning environments. Narrative-centered learning environments, which are game-based learning environments where learning activities play out in dynamically generated interactive narratives, afford great opportunity for exploring computational models of student affect. The project will explore student affect modeling through the design, implementation, and evaluation of affect models for game-based learning environments. It will develop 1) affect recognition technologies to predict students' affect states, recognize engagement and flow, and detect frustration, and 2) affect expression technologies to customize pedagogical activities and dynamically plan the empathetic responses of the virtual agents in the learning environment. All design, implementation, and evaluation activities will be carried out in Crystal Island, a narrative-centered game-based learning environment for biology.

The project will determine precisely which affect modeling techniques best close the 'affective loop' and contribute most effectively to student learning effectiveness and motivation. It is expected that the project will have a significant impact on the theory and practice of educational technology. Because of the multidisciplinary nature of the research objectives, the project will produce significant advances in computational models of affect recognition and affect expression. It is anticipated that the resulting computational models of student affect and the cognitive account of affect-informed interaction in game-based learning environments will create new learning environment technologies that promote high levels of achievement and find broad application in science education.

The project designs and implements technologies that combine artificial intelligence in the form of intelligent tutoring systems with multimedia interfaces to support children in grades 4-5 learning science. The students use LEONARDO's intelligent virtual science notebooks to create and experiment with interactive models of physical phenomena. With this technology, students' models 'come alive' as interactive multimedia artifacts that combine animation, sound, and narration. The curricular focus is on physical and earth sciences, and the technology supports multimodal interactive scientific modeling for four curricular units: forces and motion, magnetism and electricity, landforms, and weather and climate. A central feature of this environment is PadMates, which are intelligent virtual tutors that support science learning through interactive scientific modeling.

The PIs investigate the cognitive mechanisms by which learning occurs. Specifically, they study the central issues of problem solving (strategy use, divergent thinking, and collaboration) and engagement (motivation, situational interest, presence) with respect to achievement as measured by both science content knowledge and transfer. With diverse student populations in 60 classrooms drawn from both urban and rural settings, the studies determine precisely which technologies and conditions contribute most effectively to learning processes and outcomes.

The products include technologies and findings that should be the basis of a framework to inform the future development of similar systems. The impact should be substantial on all learners given the potential power of the technology to scaffold learning at an important developmental stage.

One-on-one human tutoring is remarkably effective. Seminal studies have shown that tutoring is significantly more effective than group instruction and may provide unparalleled opportunities for learning. A central, unanswered research question is, "How do expert tutors provide effective cognitive and motivational support over the course of long-term tutorial interactions to improve learning?" With a curricular focus of college-level computer science education, this project will see the design and evaluation of a computer-based intelligent tutoring system, JavaTutor, which leverages artificial intelligence to provide both cognitive and motivational support. The project will be conducted at North Carolina State University in conjunction with three partner institutions: Meredith College, Shaw University, and St. Augustine's College.

The project has three major thrusts. First, the research team will conduct a semester-long observational study of cognitive and affective tutorial support provided by expert human tutors interacting with students in a fully-instrumented online tutoring environment. The environment will log all tutorial conversations, problem-solving traces, and affective data streams including physiological signals, posture, and facial expressions. Second, the research team will develop an empirically grounded, integrated model of cognitive and affective scaffolding using machine learning techniques including hidden Markov modeling. Third, they will validate the integrated model of cognitive and affective scaffolding in a semester-long experiment with the JavaTutor intelligent tutoring system. Four versions of the JavaTutor system will be deployed and compared. It is hypothesized that over the course of a semester, the version with an integrated model of cognitive and motivational scaffolding will outperform each of the other models on both cognitive and affective student outcomes and yield differential effects across learner groups, accruing particularly significant benefit to low-performing and female students.

The products of this project include findings and technologies that will inform the future development of intelligent tutoring systems. By promoting rich learning interactions through integrated cognitive and motivational scaffolding, the project will create new learning environment technologies that promote high levels of achievement and find broad application in STEM education. It is anticipated that the resulting intelligent tutoring system technologies will serve as a foundation for the next generation of educational software that both complements and expands the impact of classroom teachers. The impact should be significant given the effectiveness of human tutoring and the potential power of these new technologies to support learning.

This project will develop a prototype intelligent cyberlearning platform for middle school audiences at a museum location to test and evaluate the use of virtual learning technologies. The content for this test is focused on sustainability issues that enable students to develop an age-appropriate understanding of the relationships between specific conservation decisions, energy use, human health, and population growth within Earth's ecosystem. The prototype cyberlearning system will demonstrate how users can learn about science topics by interacting with a display of environmental factors that enable them to explore the impact of social, economic, and technological forces that may change one existing state and condition to another. The system will enable users to understand the interrelationships of those elements by enabling them to change conditions and then observing the effect of the changes they make on the conditions presented in the initial model.


The prototype intelligent cyberlearning system will provide a unique integration of a sophisticated agent-based modeling simulation of environmental, social, and economic phenomena with three advanced learning technologies: game-based learning systems, intelligent tutoring systems, and narrative-centered learning systems. The game-based and narrative aspects of the project are embodied in the interactive time-travel focus of the 3D display on a multi-touch surface computing table in which users will play the role of environmental scientists who have been charged with helping earth become a thriving green planet. They will go back in time and be given the opportunity to make different decisions on any range of options. After they make their decisions, they will travel forward in time to see the results of their decisions. All of the interactions will be used to dynamically generate their time-travel adventures. The intelligent tutoring system will track user's problem-solving activities in the simulated world. As users make decisions, the intelligent tutoring system will draw inferences about their level of understanding of key environmental concepts. Given the current problem-solving goal (e.g., reduce green house gases) and the current state of the environment (e.g., climatological state, earth's population, factory emissions), the intelligent tutoring system will draw on its knowledge of common environmental misconceptions to assist students as they progress through the sustainability narratives. The intelligent tutoring system will receive the updated state from the agent-based simulation, which will then provide explanatory commentary and advice through the virtual human to the users about the causal connections underlying the results of the decisions they have made. Similarly, during the course of decision-making, users will be able to request advice, and the same computational framework will drive the virtual human's advice generation functionalities.


The project will design, development, deploy, and evaluate a prototype intelligent cyberlearning platform for sustainability that supports independent, but guided, exploration of science topics. Because all users interactions will be accompanied by a virtual environmental scientist who will narrate their journeys and offer problem-solving advice, users will be afforded rich learning opportunities that support independent inquiry but also provided guided exploration of complex science topics. With a focus on group learning experiences in the out-of-school setting, the virtual environmental scientist will answer questions that will engage groups of users in a collaborative effort to understand the rich interrelationships of sustainability. The project will demonstrate the transformative potential of intelligent cyberlearning systems that integrate agent-based modeling with game-based learning, intelligent tutoring systems, and narrative-centered learning in an out-of-school setting to enable users to experience science in fundamentally new ways.

North Carolina State University proposes the ENGAGE project to develop, implement, and evaluate a middle grade version of the CS Principles course that is fully situated within an immersive game-based learning environment. ENGAGE has three thrusts: (1) developing and implementing a highly engaging game-based learning environment that delivers a CS Principles course for middle grade students and their teachers; (2) making a significant educational research contribution by evaluating the effectiveness of game-based learning for computer science education; and (3) making a significant contribution to broadening the participation in computing by evaluating the effectiveness of the game-based learning framework for student learning, computing interest, and self-efficacy, particularly for underrepresented groups.

ENGAGE leverages the strengths of game-based learning, particularly narrative-centered learning, in which the game is driven by an engaging and highly motivating story. Within ENGAGE, students explore and solve global computing problems they encounter on a remote volcanic island, Crystal Island. Students work collaboratively, think creatively, deal with "big data," and analyze how their computing artifacts or those created by others are likely to impact their world. The project also includes a major focus on in-service professional development through the EngageToTeach summer institute designed not only to build teachers' computational fluency, but to instill a sense of advocacy toward computing education.

ENGAGE will see the development of a game-based learning environment for middle grade computational fluency that is expected to be highly effective for all students. With a specific focus on serving underrepresented groups, the project will draw on a partnership with North Carolina State University and diverse middle schools, including a public middle school of Johnston County, North Carolina, and a private middle school of Durham, North Carolina.

The majority of morbidity and mortality during adolescence is preventable and related to behaviors such as substance use and vehicle-related injuries. Most adolescents visit a healthcare provider once a year, providing an ideal opportunity to integrate behavioral health screening into clinical care. Although the majority of adolescent health problems are amenable to behavioral intervention, few health information technology interventions have been integrated into adolescent care. With complementary theoretical advances (social-cognitive theories of behavior change) and technology advances (intelligent narrative-centered learning environments, user modeling, and machine learning), the field is now well positioned to design health behavior change systems that can realize significant impacts on behavior change for adolescent preventive health.

Computationally-enabled models of behavior change hold significant promise for adolescent healthcare. The objective of the proposed research is to design, implement, and investigate INSPIRE, a self-adaptive personalized behavior change system for adolescent preventive health. INSPIRE will utilize a social-cognitive theory of behavior change built around a tight feedback loop in which a narrative-centered behavior change environment will produce improved behaviors in patients, and the resulting patient outcome data will be used by a reinforcement learning optimization system to learn refined computational behavior change models. With a focus on risky behaviors and an emphasis on substance use, adolescents will interact with INSPIRE to develop an experiential understanding of the dynamics and consequences of their substance use decisions. A unique feature of INSPIRE afforded by recent advances in machine learning will be its ability to optimize health behavior change at both the individual and population levels. At the individual level, INSPIRE will utilize a patient behavior model to personalize its behavior change narratives for individual adolescents. It will customize interactions based on an adolescent's goals and affective models. At the population level, INSPIRE will utilize reinforcement learning to adapt its narrative generation system to systematically increase its ability to improve two types of outcomes: behavior change and self-efficacy. The project will culminate with an experiment conducted with a fully implemented version of INSPIRE at outpatient clinics within the UC San Francisco Department of Pediatrics, Benioff Children's Hospital.

It is anticipated that INSPIRE interventions will yield two types of outcomes: 1) improved health behavior through significant reductions in adolescent risky behavior, relative to standard of care; and 2) increased self-efficacy with respect to adolescents' ability to make good decisions about their health behaviors, relative to standard of care. Designed for natural integration into clinic workflow, interoperability with EHR and patient portal systems, and security and privacy requirements, INSPIRE will report patient behavior change summaries to healthcare providers. Through multi-platform deployments supporting laptop, desktop, tablet, and mobile computing devices, INSPIRE will serve as an empowering tool for adolescents, making them full participants in their own wellbeing. It will also enable researchers to run behavior analytics to investigate which properties of alternate interventions contribute most effectively to behavior change outcomes. Going forward, it is anticipated that INSPIRE will provide a testbed for a broad range of behavior change research and serve as the foundation for next-generation personalized preventive healthcare through computationally-enabled behavior change.

Affect, or emotion, profoundly shapes human experience. It influences how we perform tasks, how we build relationships with one another, and how we navigate the complexities of our daily lives. Affect is shaped and influenced by communication with other humans, experiences with the natural world, and interactions with machines. Affect plays a particularly prominent role in learning. During learning, a recurring subset of the broad range of human emotions such as confusion, frustration, boredom, anxiety, engagement, surprise, and delight appear regularly. Different emotions are best responded to in different ways. For example, task-based feedback and guidance is a helpful response to emotions of confusion and frustration, while empathetic feedback is more helpful for emotions of anger or excitement. Prior research has not answered the question of how affective adaptation can maximize the benefit to students as they interact with interactive computer-based learning environments. And yet the investigators on this project are now well positioned to address a central, unanswered question of how learning environments can adaptively respond to students' affect to create the most effective, engaging learning experiences while simultaneously promoting improved attitudes toward learning.

The project will provide important societal benefits by generating theoretical and practical advances across multiple disciplines. The project will lead to a deeper understanding of affect-rich learning; a set of broadly applicable affect adaptation principles; and a computational model of affective adaptation and dialogue that will be incorporated into a learning environment for science learning. The resulting affect-modeling technologies can serve as a foundation for the next generation of adaptive educational software that will promote learning through affect-rich adaptation. This will be broadly useful throughout education. The project will address issues of diversity by partnering with the highly diverse Dunn Middle School and Harnett Central Middle School, and through ongoing collaboration with the STARS Alliance for Broadening Participation in Computing. To ensure societal impact, the results will be disseminated to the public through middle school outreach programs, and to the scientific community through publication at scientific venues.

The three major scientific goals of the project are to: (1) Capture rich multimodal data of students' affective experiences while interacting with a fully instrumented learning environment with spoken dialogue. Observational studies will be conducted by having middle school students interact with an existing learning environment for science education called "Crystal Island." Crystal Island was developed by the investigators on this project and has already been used by thousands of students in middle school classrooms to learn microbiology, but it does not currently support rich multimodal interaction or natural language dialogue. Crystal Island will be fully instrumented to collect rich, multimodal data including speech, facial expression, gaze, posture, skin conductance response, heart rate, and problem-solving actions. (2) Design, develop, and refine an affect-understanding model that integrates students' natural language, nonverbal behavior, physiological response, and task-action phenomena into a rich multi-dimensional stream of affective data. By utilizing this data collected from the observational studies, an affect-understanding model will be constructed using machine learned including hidden Markov modeling. This will be the first affect-understanding model for learning environments that integrates the full complement of affect signals of spoken language (including prosody, syntax, and semantics), nonverbal behavior (including gaze and posture), physiological data (including skin conductance response and heart rate), and task actions (including navigation and manipulation actions in the learning environment). (3) Design, develop, and refine an integrated affect and dialogue management model that adaptively responds to students' affective states in the course of their learning interactions. By utilizing the learning-interaction data collected in the observational studies, a Partially Observable Markov Decision Process (POMDP) affect adaptation policy will be acquired with reinforcement learning, integrating affect and dialogue management. The resulting adaptation policy will govern both when and how the system responds to students' affect as they solve problems. The computer-based mentor will provide problem-solving advice, encouragement, empathetic responses, and other support as is needed to improve the educational experience and outcome.

The investigators will research how characteristics of intelligent virtual humans (IVHs) support the ability of students to reflect on and, therefore, improve their learning in undergraduate biology. To date, research has shown mixed effectiveness when human avatars are used in learning technologies. To remedy that, the researchers will first study how expert human tutors use verbal and facial cues in reacting to students' cognitive, affective, metacognitive, and motivational (CAMM) processes. Then, they will use these data to build an enhanced intelligent virtual human tutor (by altering software called "MetaTutor"). The project will advance the field's ability to build more effective intelligent tutors and advance understanding of self-regulated learning.

The researchers propose to experimentally study the effectiveness of the enhanced IVHs on learners' self-regulatory processes and other learning outcomes. Data will be collected on both a natural face and a natural face that has been morphed and presented as a virtual human. The facial and verbal expressions are meant to provide learners with an additional information source they can use to monitor and regulate their ongoing self-regulatory processes, including making accurate emotional appraisals. In addition to the facial data, the researchers will collect self-report data, trace data using a variety of sensors, learning outcomes (e.g., pretest and posttest), and knowledge construction activities (e.g., summaries of content, notes, quizzes). Finally, the project will be disseminated in the form of journal publications, conference presentations, and an enhanced version of MetaTutor.

The STEM+Computing Partnership (STEM+C) program seeks to advance multidisciplinary integration of computing in STEM teaching and learning through applied research and development across one or more domains. Building on a previously developed and tested prototype for computer science education in a middle school oceanography elective, the ENGAGE project's objective is to design and study a more comprehensive curricular strategy for middle school life science using a game-based learning environment that will deeply integrate computational thinking practices within the life science curriculum. Topics will include the study of interdependent relationships of species in ecosystems, matter and energy in organisms and ecosystems, and natural selection of species and their adaptations for survival. Along with a focus on student learning, the project will develop and study a teacher professional development model to support teachers in the integration of computational thinking in middle school life science. Principal Investigators will gather specific evidence about the ways in which game-based learning can effectively support computationally rich science practices aligned with the new Science Education Framework (National Research Council, 2012) and the Computer Science Teachers' Association, Computer Science Education Computational Thinking Framework (2016). Specifically, the effort will: (1) design an innovative curricular strategy and novel game-based learning environment to develop computationally rich science practices (developing and using models, and analyzing and interpreting data), for middle school students from North Carolina and Florida; (2) investigate how middle school students develop computational thinking practices (creating abstractions and models, analyzing problems and artifacts, and developing computational artifacts) in middle school science classrooms with the game-based learning environment; and (3) develop an evidence-based teacher professional development program that supports teachers in the deep integration of computational thinking into middle grades science.

The overarching research question will be: How can deep, mastery-oriented gameplay develop core computational thinking practices in middle school life science? Related sub-questions will include: (a) How can game-based learning promote the development of students' science and engineering practices?; (b) How can game-based learning promote student mastery of computational thinking deeply infused science education?; and (c) How can game-based learning for computationally rich science improve students' attitudes in science with regard to short-term self-efficacy, outcome expectancy, and long-term STEM career interest? The project will undertake three types of activities carried out in parallel across each year of the project. First, it will develop and iteratively refine the curricular strategy and collaborative game-based learning environment, which will be implemented at partner middle schools in North Carolina and Florida, with increasing numbers of teachers and students participating in each succeeding year. Second, the study will develop, implement, and refine the corresponding professional development materials and associated online and face-to-face activities. The teacher professional development will be implemented throughout the school year, utilizing both a summer institute and ongoing school-based support. Third, classroom studies will be conducted to build the evidence base on how middle school students can most effectively develop computationally rich science practices. For the classroom studies, the project will compare three conditions: (1) the baseline condition (standard science classroom practice with no project implementation) will utilize teachers who have not previously participated in the project; (2) the project-without-game condition, where students will participate in an implementation of the curricular strategy that provides out-of-game computational problem solving on data-rich science problems outside of the game, but does not provide the game-based learning experiences; and (3) implementation of a full version of the ENGAGE curricular strategy that provides both game-based learning experiences, as well as outside-of-game computational problem solving. Data-gathering strategies throughout the three-year duration of the study will include: classroom observations, individual teacher interviews, student and teacher focus groups, project-developed assessments, embedded game-based learning assessments, student-created artifact assessments, pre- and posttest assessments, and assessments of science and STEM attitudes. Data interpretation strategies will include: comparative analysis of scores across years and across the three experimental conditions, observational and formative analyses to guide refinement of learning tasks, qualitative and quantitative analyses of student performance and growth, and formative analysis of changes in self-reported attitudes. The main outcome of the project will be a research-informed and field-tested prototype integrating computational thinking into science learning at the middle school level. An external organization will conduct the formative and summative evaluations.

The "Engaging STEM Undergraduate Students in Computer Science with Intelligent Tutoring Systems" project will see the design, development, and evaluation of the Personalized Real-time Intelligent Mentoring Environment (PRIME) system, which will provide adaptive problem-solving support and adaptive motivational support to help students learn to solve computing problems. PRIME will track each student's progress while providing real-time feedback, multiple levels of hints, and customized problem-solving advice throughout students' learning interactions. The project will use the block programming language "Snap!" as it eliminates most syntax errors, freeing students to focus on concepts and logic.

Introductory computing is a required component of many STEM undergraduate curricula, and because of the fundamental importance of computing for all STEM disciplines, courses that introduce computing to STEM undergraduates hold enormous potential for shaping the way students develop computational fluency that will serve them throughout their careers. The PRIME intelligent tutoring system will be evaluated in a large introductory computing course at North Carolina State University, which serves more than 2,000 students each year. The PRIME environment will also be evaluated at the University of Florida and at Florida A&M University, a Historically Black University. Experimental studies with PRIME will have a dual focus: investigating its impacts on helping students learn about computing (analyzing problems, creating models and abstractions, and building and refining programs) and improving students' attitudes towards computing (self-efficacy for computing and interest in computing).

This is a research project supporting a new model of Computer Supported Collaborative Learning (CSCL) that combines the advantages of game based learning with problem based learning. Good game based learning environments combine rich scenarios with engaging activities to serendipitously provide student learning. These learning environments also provide an opportunity for players to collaborate in reaching their game goals. Good problem based learning environments provide support for the solution of complex and ill-structured problems. The combination of these two types of learning environments promise to provide the engagement and richness of game based learning with the support environment to engage students in authentic science. Both of these environments are computer based so the actions and interactions of the students and teachers are captured for analysis. Applying learning analytics to the captured data provides information on student learning for the teacher, provides learning information to the student for self-reflection and improved learning, and provides information for the system designer to improve the effectiveness of the new CSCL environment.

The scientific problem domain is environmental science for middle school students. The CSCL environment is a game based learning environment that incorporates problem based learning. The interaction between the CSCL environment and the student is enhanced by the collection of data on the student based on cognitive, affective, and metacognitive states that are inferred using artificial intelligence technologies. Specific strategies are employed to help students construct explanagions, reason effectively, and become self-directed learners. Key outcomes of the project include a model of collaborative scaffolding for game based learning that is usable in classrooms to help students learn STEM content and learning analytics designed to support the teacher in the roles of guide and collaborator. A goal of the project is wide dissemination of the CSCL system.

The project is supported by the Education and Human Resources Core Research program, which supports fundamental research in STEM learning and learning environments. Reflection plays a critical role in student learning and encompasses a broad range of cognitive and metacognitive processes enabling students to (1) think critically about their learning processes, (2) integrate new information with prior knowledge, (3) form and adapt learning strategies, (4) view concepts and knowledge from multiple perspectives, (5) generate self-explanations to enrich conceptual understanding, (6) compare learning processes and artifacts to those created by experts and peers, and (7) make metacognitive judgments about knowledge. The National Research Council recently concluded that systematic reflection is essential for deep learning and effective educational practice. This project investigates a suite of theoretically grounded, adaptive game-based reflection tools to scaffold students' cognitive and metacognitive reflection with the overarching objective of improving middle school students' science problem-solving processes. These reflection tools are integrated as part of the existing Crystal Island learning environment.

In studying middle school students' cognitive and metacognitive reflective processes, three key research questions will be addressed: 1) How do embedded and retrospective reflection scaffolds promote improved learning outcomes including science problem-solving skills, science content knowledge, metacognitive awareness, and reflection skills? 2) How can learning analytics be leveraged to extend models of reflection and self-regulated learning for problem solving within game-based learning environments? and 3) How can we create adaptive scaffolding for game-based learning environments that best foster reflection during and following science problem solving? The project builds upon a theoretical framework of self-regulated learning to inform the development of embedded and retrospective scaffold tools, gathering both process data (e.g., log data, eye tracking, think-aloud) and outcome data (e.g., pre- post- learning, transfer, metacognition, self-efficacy) to inform design principles and descriptive models of the role of reflection in students' problem-solving processes. Key project outcomes include an empirically grounded theoretical framework for reflection-enhanced learning and learning analytic techniques that yield predictive models of reflection in science problem solving.

Project Summary/Abstract Engaging adolescents' interest in pursuing careers in health science and the health professions offers significant promise for building our nation's healthcare and health research capacity. The goal of this project is to create Health Quest, an intelligent game-based learning environment that increases adolescents' knowledge of, interest in, and self-efficacy to pursue health science careers. Three specific aims will be accomplished by the project: 1. Design and develop Health Quest to engage adolescents' interest in the health sciences utilizing personalized learning technologies that integrate the following components: (a) the Health Quest Career Adventure Game, an intelligent game-based learning environment that leverages AI technologies to create personalized health career adventures; (b) the Health Quest Student Discovery website, which will feature interactive video interviews with health professionals about their biomedical, behavioral, and clinical research careers; and (c) the Health Quest Teacher Resource Center website, which will provide online professional development materials and in-class support for teachers' classroom implementation of Health Quest. 2. Investigate the impact of Health Quest on adolescents' (1) knowledge of biomedical, behavioral, and clinical research careers; (2) interest in biomedical, behavioral, and clinical research careers; and (3) self- efficacy for pursuing biomedical, behavioral, and clinical research careers by conducting a matched comparison study in middle school classes. ! 3. Examine the effect of Health Quest on diverse adolescents by gender and racial/ethnicity. Working closely with underrepresented minorities throughout all design and development phases of the project, the project team will specifically design Health Quest to develop girls' and members of underrepresented groups' knowledge of, interest in, and self-efficacy to pursue health science careers.

The Future of Work at the Human-Technology Frontier (FW-HTF) is one of 10 new Big Ideas for Future Investment announced by NSF. The FW-HTF cross-directorate program aims to respond to the challenges and opportunities of the changing landscape of jobs and work by supporting convergent research. This award fulfills part of that aim.

K-12 STEM teachers are critical to the US economy. With investments in teaching quality representing enormous economic value, the quality of education has been identified as a significant determinant of gross domestic product economic growth. However, the US teacher workforce is experiencing a crisis: teacher demand exceeds supply at every level, and attrition is extraordinarily high for new teachers. Further, while STEM teaching represents one of the areas of highest need, STEM teachers leave the profession at high rates. These developments call for innovative workforce augmentation technologies to improve K-12 STEM teachers' performance and quality of work-life. To address this critical national need, the project will investigate how intelligent cognitive assistants for teachers can transform teacher work to significantly increase teacher performance and teacher quality of work-life. The project centers on the design, development, and evaluation of the Intelligent Augmented Cognition for Teaching (I-ACT) framework for intelligent cognitive assistants for teachers. With a focus on assisting K-12 STEM teachers in technology-rich inquiry teaching that supports collaborative, problem-based STEM learning, I-ACT cognitive assistants provide teachers with (1) prospective pedagogical guidance (preparation support preceding classroom teaching), (2) concurrent pedagogical guidance (real-time support during classroom teaching), and (3) retrospective pedagogical guidance (reflection support within a community of practice following classroom teaching). The project will culminate with an experiment conducted with a fully implemented version of I-ACT in public middle schools in North Carolina and Indiana.

The project realizes its objective through two primary thrusts. First, the research team will design and develop I-ACT cognitive assistants for K-12 STEM teachers and test them in public school classrooms. Utilizing AI-based multimodal learning analytics and a social constructivist theory of pedagogy, I-ACT cognitive assistants use machine-learned models of teacher orchestration to provide guidance throughout the full teaching workflow. I-ACT cognitive assistants operate in a tight feedback loop in which collected data will drive successive iterations of machine learning to train refined teacher support models for improved I-ACT cognitive assistant functionalities. Second, the research team will investigate how I-ACT cognitive assistants improve K-12 STEM teacher performance and teacher quality of work-life. The team will conduct focus groups, case studies, semi-structured interviews, and observations of teachers using I-ACT cognitive assistants in school implementations with middle school science teachers at the project's partner schools. The team will also conduct quasi-experimental studies to determine I-ACT impact on teacher performance and quality of work-life.

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.

This project aims to serve the national interest by creating ExplainIt, an innovative explanation-based classroom response system that provides real-time support to undergraduate STEM students and instructors by using natural language processing to analyze student explanations of STEM phenomena. The project will advance understanding of how to improve undergraduate STEM education by providing real-time formative feedback to each individual student and real-time summaries to instructors so they can quickly adapt their instruction to the current needs of their students. The project will produce significant theoretical and practical advances in undergraduate STEM education. It will lead to a deeper understanding of how students learn with explanation-based classroom response systems, including the learning gains and improvements in student engagement. It will also lead to a set of effective instructional support principles for explanation-rich classroom interactions that will be broadly applicable in multiple STEM disciplines and in diverse institutional settings. Together, these advances will yield fundamental improvements in undergraduate STEM education.

Three goals guide this project. First is to create the ExplainIt explanation-based classroom response system. Second is to produce empirically grounded research results on adaptively supporting student learning and engagement with ExplainIt. Third is to develop online resources that enable instructors to easily integrate ExplainIt into their teaching. Each year, instructors in biology, computer science, and physics will contribute to the design, development, and evaluation of the ExplainIt software. The project will investigate conditions under which improved student learning occurs by evaluating ExplainIt in a wide range of undergraduate STEM courses. ExplainIt will be evaluated with more than 6,000 students in introductory courses, lower-division courses, and upper-division courses. The evaluation will have a dual focus: evaluating ExplainIt’s effects on students’ STEM learning (conceptual knowledge, problem solving), and evaluating ExplainIt’s effects on improving students’ engagement in STEM (STEM self-efficacy, STEM interest). The project team will grow the ExplainIt community of practice through workshops and presentations at conferences on biology education, computer science education, and physics education, as well as through webinars and social media. The NSF IUSE: EHR Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools.

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.

Emerging breakthroughs in AI create significant and timely opportunities for accelerating innovation relevant to pressing national challenges in STEM education. Advances in AI are enabling new levels of interactivity, engagement, and inclusion. Driven by a vision in which AI supports and extends the intelligence of teachers and learners, the NSF AI Institute for Engaged Learning will design, develop, and investigate AI-driven narrative learning environments that create engaging story-based, collaborative problem-solving experiences. The Institute will conduct foundational AI research in natural language processing, computer vision, and machine learning, as well as in AI ethics. Building on these foundational advances, the Institute will conduct use-inspired AI research on narrative learning environments with rich AI-driven virtual agents and powerful multimodal sensing capabilities to understand how students learn and collaborate in rich story-based problem scenarios. The Institute will provide a robust infrastructure to support at-scale implementations of AI-driven narrative learning environments. It will create a nexus for distinctive innovations in in-school and out-of-school STEM education. Its research vision will be to empower diverse learners to become the next-generation STEM workforce by creating generative, collaborative AI-driven narrative learning environments that deeply engage learners in schools, at museums, and within their own communities. This vision will be informed by connections with diverse stakeholders to ensure that the Institute’s learning environments are ethically designed and promote diversity, equity, and inclusion.

The NSF AI Institute for Engaged Learning will produce transformative advances in STEM teaching and learning by bringing together a team of diverse researchers from four universities (North Carolina State University; Indiana University, University of North Carolina at Chapel Hill; Vanderbilt University), as well as an educational non-profit organization (Digital Promise) which will bring educational practitioners, policy makers, and product developers into the work. The Institute’s partners include a national network of K-12 schools, museums, and non-profit organizations. The Institute will create narrative learning environments that generate interactive stories dynamically tailored to the needs and interests of individual students and small groups and in multiple settings (classrooms, after-school programs, and museums). The Institute’s research has three complementary thrusts, to create: (1) narrative learning environments that generate engaging interactive story-based problem scenarios that elicit rich communication, require coordination, and spark collaborative creativity; (2) embodied conversational agent technologies with multiple modalities for communication (speech, facial expression, gesture, gaze, and posture) to support engaging learning interactions. Embodied conversational agents will be driven by advances in natural language understanding, natural language generation, and computer vision; and, (3) an innovative multimodal learning analytics framework that analyzes parallel streams of multimodal data derived from students’ conversations, gaze, facial expressions, gesture, and posture as they interact with each other, with teachers, and with embodied conversational agents. Woven throughout the Institute’s activities will be a strong focus on ethics, with an emphasis on creating AI-augmented learning that is deeply informed by considerations of fairness, accountability, transparency, trust, and privacy.

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.