United States

Research shows that effective teachers are the most important factor contributing to student achievement. Although curricula, reduced class size, funding, family, and community involvement all contribute to school improvement and student achievement, the most influential factor in the classroom is the teacher. Yet it has become clear that the professional development needed to support teachers and how effectively they function within classrooms is often lacking or ineffective. In many parts of the globe, teachers who need the most professional development (e.g. new or underqualified) often receive the least. Internationally, there is also a recognized need for improved instruments and methodologies to gauge elements of classroom quality and effective teaching. Particularly in low-income and fragile contexts, where many teaching personnel are underprepared and under resourced, more rigorously developed and culturally attuned observation tools have the potential to provide much needed feedback to teachers in a continuous cycle of improvement. In our previous RLO grant (Toward the Development of a Rigorous and Practical Classroom Observation Tool: The Uganda secondary school project), we developed and validated the Teacher Instructional Practices and Processes System (TIPPS) with learning outcomes in secondary schools in Uganda. Using this observational tool, we examined the quality of teaching practices and classroom processes through live observations. Subsequently, we developed a pre-school version of the TIPPS in Ghana that was found to have meaningful associations with both learning and socio-emotional outcomes. We also piloted a primary school version of the TIPPS in India, where an NGO is using the TIPPS as a guide to provide teacher feedback. To date, we have not had the opportunity to systematically employ TIPPS as a feedback tool in a supportive fashion to improve teaching practices, student learning and teacher outcomes. Creating this cycle of continuous improvement is the goal of the present investigation, albeit in a new cultural context. To test this in the Honduran context is an idea that grew organically, thanks in large part to the yearly gatherings of RLO colleagues that allowed for cross-pollination of ideas and discussions on topics of interest. Since our first RLO meeting in London, we have been speaking with Erin Murphy-Graham and her team about how we could join forces to augment the impact of the Sistema de Aprendizaje Tutorial program (Tutorial Learning System or SAT) in Honduras. The current proposal represents one-half of two parallel, collaborative but separate investigations. Murphy-Graham's proposal seeks a deeper understanding of which SAT pedagogical practices are effective (as assessed by the TIPPS) in impacting student learning and social and emotional outcomes, as well as how pedagogical practices effect teacher motivation by examining an intensive SAT condition without the addition of feedback. In this way, she hopes to recommend improvements to her partners in the SAT program. Our parallel proposal, which would operate in tandem with her existing intervention work, serves to further our objectives to both extend TIPPS' cross-cultural reach and systematically test its use as a feedback tool in the context of an optimally supportive structure. Using the SAT programming, we seek to develop an empirically-based, robust feedback mechanism that cultivates improvement in a continuous cycle of change

Children in conflict-affected countries (CACs) experience profound constraints on their academic learning and socioemotional well-being. Children exposed to violence and poverty come to "school" (formal or non-formal education settings) with poor executive function skills (e.g. working memory, inhibition, attention), emotional/behavioral regulation skills and social-information-processing skills. And the formal and non-formal "schools" they attend rarely use effective strategies to advance their academic and socioemotional skills. What can be done? This project aims to develop both scientific and practical knowledge about how to lift these constraints on children in CACs like Niger. First, we propose to adapt and test novel, low-cost, targeted interventions (LCTs) like Mindfulness (MI) and Brain Games (BG) designed to improve children's executive function, emotional/behavioral regulation and social information-processing skills and, subsequently, their literacy and numeracy skills and socioemotional well-being. Second, even when interventions like MI and BG work, they often work better for some children, in some classrooms and schools, and under some conditions than for others. So this study will examine whether variability in the quality of implementation (e.g. dosage, fidelity) of MI and BG results in the variability in their impacts on children's learning and development in Niger. Third and finally, these types of complex interactions among students, teachers, "schools" and program interventions (like remedial support programs) are embedded in larger systems and broader contexts that may constrain or enable quality implementation of program strategies (such as MI and BG). But there is very little high-quality, rigorous research, grounded in social and systems theories, available in CACs to understand how these higher-level systems influence the dynamic interactions of schools, programs, classrooms, teachers and students. So we will conduct a theory-building qualitative study embedded in the school cluster-randomized field test of MI and BG and their implementation. Through this project, we hope to (a) achieve a dynamic, multi-level understanding of efforts to improve learning processes and outcomes for refugee, IDP, and local children in Niger and other CACs; (b) contribute to the synthesis of the developmental, educational, prevention and social sciences in theory and method; and (c) have a catalytic effect on the education in emergencies sector by identifying effective, scalable strategies that improve children's learning and development.

We propose to develop an AI-based approach to modelling the functional organisation of the human motor system and, in collaboration with our international partners, apply this methodology to provide important insights into motor control across the lifespan. This project is designed around three key objectives that address current methodological limitations and research gaps in our mechanistic understanding of healthy neuromotor development and decline. Our first objective in developing this AI-based approach is to provide the motor control community with a new toolbox for the modelling of motor coordination. Despite recent theoretical developments in the neural organisation of movement generation, analytical methodologies for testing these theories in neurophysiological and motor-behavioural data are currently missing. Here we will address this gap by developing a novel approach to analysing and modelling the output of neuromotor circuitry in the form of muscular interactions ('muscle synergies'). Crucially, our proposed framework will enable a characterisation of the functional roles of muscle interactions with respect to task performance, introducing a novel perspective to the concept of muscle synergy. Building on our previous work, we will develop novel information-theoretic measures and combine them with AI tools from machine-learning and statistical physics to quantify functionally diverse muscle interactions that are necessary for task performance. This innovation will represent a paradigm shift in analytical approaches that align with the forefront in theoretical understanding on human movement modularity. For our second objective, we will partner with colleagues in University of Paris-Saclay (France) to provide comprehensive insights into the developmental origins of movement modularity. Together, we will address a current research gap in mechanistically understanding motor development by applying the framework to new-born infants analysed longitudinally from early kicking postpartum to independent ambulation. This partnership represents an important opportunity to delve into the elusive origins of behavioural flexibility in early life through this novel perspective on muscle synergies. Specifically, we will characterise motor development in a way akin to the development of language (i.e. constructing movements by combining simple building-blocks of movement in progressively more complex ways). For our third objective, we will work with colleagues in New York University (USA) to apply the framework to data on healthy older adults performing activities-of-daily-living. This partnership will significantly improve the mechanistic understanding of senescence on motor behaviour. During this collaboration, we will further develop our framework towards a crucial frontier for the field and practical applications in the effective mapping of brain-muscle interactions underlying everyday movement. By applying it across younger and older adults, we will holistically capture the neuromotor interactions underlying everyday movements and how they are affected by aging. As movement is central to the functional independence of the elderly, our work will inform parallel lines of aging research in the UK. Furthermore, understanding the effect of healthy aging on motor coordination will serve as benchmark against which impairments as a result of age-related diseases can be compared, enabling the discrimination of healthy from pathological decline in motor function. Ultimately, by comparing our findings across the three age groups (infants, younger adults and older adults), we will characterise the evolution of muscle synergies from early life to older age. In sum, our project proposes a novel perspective to the study of motor coordination via the development of an AI framework to progress our understanding of human action across the lifespan.

As a mathematician working in the field of statistical mechanics, I am pursuing research problems that concern mathematical models that exemplify some important aspect of the physical world. The aim is find the simplest viable description of the typical large scale behaviour of such systems. For example, there are huge number of air molecules in a room, so that, microscopically, the behaviour of the air in the room is described by all of the positions and velocities of those molecules. This is a vast amount of information. In the large, however, what effectively describes the air is a few parameters, such as the temperature and the pressure, that are specified by averages of particles over small regions of space (that nonetheless contain very many particles). In the rigorous theory of statistical mechanics, we choose a suitable mathematical model of a physical system, and prove how the behaviour of such macroscopic quantities as temperature and pressure arises from the microscopic structure of the system.In this proposal, I am undertaking three related research projects, each of which reflects in some way this theme:I. Phase boundary fluctuation. If oil is injected into still water, it forms into a droplet that makes the total surface tension at the boundary as small as possible. On a finer scale, however, the boundary between the two substances may be random. In a recent series of papers, I have investigated, for a natural mathematical model of two such substances, the geometry of this random boundary. I am proposing to investigate what is universal about this random fluctuation: that is, which elements of this behaviour are shared with a diverse range of other systems. II. Trapping in disordered media.If a charged particle in an electric field moves in an environment populated by occasional obstacles, its progress is liable to be frustrated by traps formed by the obstacles. What is the geometry of the traps that waylay the particle, and to what degree do these traps slow down the walk? Alexander Fribergh and I are carrying out an extensive investigation of a mathematical model of this problem, in which a walker jumps generally in a preferred direction, but makes other random moves as well, on a grid in which some edges are impassible.III. Spatial random permutations.At very low temperatures, helium condenses into a remarkable substance that flows with extreme ease. A mathematical model of repelling random particles is naturally associated to such low-temperature gases. I am planning to investigate how these particles behave in a fashion that, while random, has large scale order, and how this order is related to the special properties of very cool gases.