36633 research outputs found
Sort by
‘Literal Torture’:Vulnerability, resilience and young people's experiences of pressure in physical education
No full textThis paper combines vulnerability and resilience theory to explore the pressure young people experience in Physical Education (PE) and sport at secondary school. The theoretical framework was used to understand both how young people experience PE in school and how vulnerability and resilience function interdependently in social contexts like schools. The research foregrounded the perspectives of the 154 pupils aged 14–16 in two secondary schools in North‐East England, using an initial focus group, a survey and individual interviews in each school. Thematic data analysis revealed causes and effects of pressure in PE, along with its sources and the factors that intensify and ameliorate it. Findings suggest that pupils who designate themselves ‘high’ ability feel pressure in PE, as well as those who regard themselves as ‘low’ ability, but it is more prevalent in, and detrimental to, the latter group. Ability level also emerged as one of the factors that intensified pressure alongside (lack of) enjoyment. Peers, individuals themselves and teachers were identified as the key sources of pressure. The paper concludes by suggesting that PE in school needs reform to alleviate this pressure, which creates a complex and often unrecognised dynamic of vulnerability and resilience in pupils. This dynamic often has negative effects on pupils regardless of ability when responsibility is assigned too much to them as individuals. We propose increasing pupils' voice and choice in school to break their association with individualised pressure, vulnerability and resilience in PE and improve young people's physical and mental health
Utilizing Large Language Models for Machine Learning Explainability
No full textThis study explores the explainability capabilities of large language models (LLMs), when employed to autonomously generate machine learning (ML) solutions. We examine two classification tasks: (i) a binary classification problem focused on predicting driver alertness states, and (ii) a multilabel classification problem based on the yeast dataset. Three state-of-the-art LLMs (i.e. OpenAI GPT, Anthropic Claude, and DeepSeek) are prompted to design training pipelines for four common classifiers: Random Forest, XGBoost, Multilayer Perceptron, and Long Short-Term Memory networks. The generated models are evaluated in terms of predictive performance (recall, precision, and F1-score)and explainability using SHAP (SHapley Additive exPlanations). Specifically, we measure Average SHAP Fidelity (Mean Squared Error between SHAP approximations and model outputs) and Average SHAP Sparsity (number of features deemed influential). The results reveal that LLMsare capable of producing effective and interpretable models, achieving high fidelity and consistent sparsity, highlighting their potential as automated tools for interpretable ML pipeline generation. Importantly, our findings indicate that the quality of LLM-generated pipelines closely approximates that of manually engineered solutions both in predictive accuracy and explainability, suggesting that LLMs can reliably support or even partially automate explainable ML workflows with minimal loss in interpretive depth
Land Actuator:Fiji Arts Council Exhibition of LAGI2025 shortlisted entries,Suva, Fiji
No full textSubmission shortlisted and selected for exhibition at the Fiji Arts Council, Suva, Fiji. Exhibition opening 6th November 2025. The Land Actuator by Maria Sol Meyer, Sarah Stevens, Hugo Mulder, Carlos Lora, and Sol Macchiavello (Argentina) incorporates semi-transparent solar photovoltaic modules, compressed earth brick cisterns, and fog harvesting to produce 135 MWh of electricity and 3.7 million liters of filtered water. A submission to LAGI 2025 Fiji, held in collaboration with the Village of Marou
Sustainable Photography?
No full textA group exhibition featuring a picture from an ongoing series in Fergus Heron’s wider work exploring how photography can reveal changing natural formations. The picture exhibited is titled Allt Dearg, Highland, Scotland, 2020. The title includes Scots Gaelic words describing the place in which exposure in camera was made. The location is considered actively productive of photography with ecological features as part of photosynthetic processes. The exhibited print uses the chromogenic process involving the projection of light onto paper to realize the finished picture for gallery viewing
Neuroplastic Reflective Game Design: A Framework Bridging Neuroscience and Game-Based Learning
No full textThis paper introduces Neuroplastic Reflective Game Design (NRGD), a theoretical framework that bridgesneuroscience and game-based learning by linking reflective gameplay to underlying neuroplastic mechanisms.Neuroplasticity, the brain’s capacity to reorganise neural connections, underpins learning, memory, and adaptability, yet its potential role in educational design remains underexplored. Reflection, defined as the deliberate evaluation of experience, engages prefrontal and cingulate regions, reinforcing executive functions and long-term retention. Digital games provide fertile ground for embedding structured reflection because they combine immersion, interactivity, and feedback, but their capacity to deliberately support neuroplasticity has not been systematically theorised. A structured literature review synthesised insights from neuroscience, education, and game design. The review revealed that while each domain offers valuable perspectives, such as neural mechanisms of plasticity, pedagogical models of reflection, and game-based scaffolds for metacognition, they remain siloed. Neuroscience often stops at describing mechanisms, education frames reflection as pedagogy without neural grounding, and games emphasise engagement without connecting to brain adaptability. To address this, the study investigates three guiding questions: (1) how reflective game design can be theoretically extended to support neuroplasticity within digital learning environments; (2) which cognitive and neural mechanisms may be activated through reflection in gameplay; and (3) what design principles can be derived to inform future interdisciplinary work. The NRGD framework responds through four cyclical phases (Gameplay, Assessing Conceptualisation, Active Experimentation in Level Up, and Reflective Feedback). Each one is mapped to cognitive functions and neural processes such as long-term potentiation, synaptogenesis, and error-driven adaptive rewiring. An illustrative example in music theory demonstrates how these phases can be operationalised in practice, and a summary table aligns design features with their associated neurocognitive outcomes. The framework offers practical value for educators and instructional designers seeking deeper learning, for game developers aiming to align mechanics with cognitive science, and for neuroscientists and clinicians exploring applications in neurorehabilitation, lifelong learning, and therapy. While conceptual in nature, the framework also identifies directions for empirical validation, methodological refinement, and adaptation across domains. By bridging pedagogy and neuroscience, NRGD establishes a novel theoretical foundation for designing digital games that are both pedagogically effective and biologically grounded.<br/
Special Issue on “Low Carbon Design and Manufacturing Process”
No full textThe design and manufacturing process account for approximately 60–80% of a product’s lifecycle carbon emissions, making them critical junctures for achieving low-carbon transformation across multiple industrial sectors [...
Prevalence of stress, anxiety, depression, and reported self-esteem among mothers of children with autism spectrum disorder
No full textBackground: Mothers of children with autism spectrum disorder (ASD) have an increased risk of depression, anxiety, stress, and diminished self-esteem due to the numerous challenges involved in raising children with socio-behavioural difficulties.Objectives: This study investigated the prevalence of depression, anxiety, stress, and low self-esteem among mothers of children with ASD.Materials and methods: A cross-sectional descriptive study was conducted among 246 mothers of children with ASD. The prevalence of depression, anxiety, and stress was assessed using the Depression, Anxiety and Stress Scale-21 Items, and self-esteem was measured using Rosenberg’s Self-Esteem Scale. Demographic data for both mothers and children were collected using a pro forma. Data are summarised as frequencies, means (M), standard deviations (SD), and percentages.Results: The prevalence rates among mothers were as follows: depression, 84.14% (N=207); anxiety, 77.2% (N=190); stress, 74% (N=167); and diminished self-esteem, 67.9% (N=167). Mean severity scores indicated moderate levels of depression (M=17.26, SD=7.7), anxiety (M=12.7, SD=7.7), and stress (M=19.49, SD=8.0), and low self-esteem (M=23.25, SD=9.5).Conclusion: Mothers of children with ASD had high prevalence rates of depression, anxiety, stress, and low self-esteem. These findings highlight the psychological challenges faced by mothers caring for children with ASD. Therapeutic interventions should therefore address the mental health needs of these mothers alongside the rehabilitation of their childre
Progressive Cultures in Interwar Britain:Situating Leslie Paul
Get PDFAn 8000-word chapter on the contributions of Leslie Paul to Woodcraft Folk, contextualised in a wider reading of his non-fiction writings and autobiographies, and by his wider organisational networks, especially the Federation of Progressive Societies and Individuals
A systematic in vitro study of the effect of normoglycaemic and hyperglycaemic conditions on the biochemical and cellular interactions of clinically-available wound dressings with different physicochemical properties
Get PDFDiabetic foot, leg ulcers and decubitus ulcers affect millions of individuals worldwide leading to poor quality of life, pain and in several cases to limb amputations. Despite the global dimension of this clinical problem, limited progress has been made in developing more efficacious wound dressings, the design of which currently focusses on wound protection and control of its exudate volume. The present in vitro study systematically analysed seven types of clinically-available wound dressings made of different biomaterial composition and engineering. Their physicochemical properties were analysed by infrared spectroscopy, swelling and evaporation tests and variable pressure scanning electron microscopy. These properties were linked to the interactions with inflammatory cells in simulated normoglycaemic and hyperglycaemic conditions. It was observed that the swelling behaviour and evaporation prevention at different glucose levels depended more on the engineering of the fibres than on the hydrophilicity and hydrophobicity of their biomaterials. Likewise, the data show that the engineering of the dressings as either non-woven or woven or knitted fibres seems to determine the swelling behaviour and interactions with inflammatory cells more than their polymer composition. Dressings presenting absorbent layers made of synthetic, non-woven fibres supported the adhesion of monocytes macrophages and stimulate the release of factors known to play a role in the chronic inflammation. Non-woven absorbent layers based on carboxymethyl cellulose mainly stimulating the iNOS, an enzyme producing free radicals; inthe case of Kerracel this was combined with a swelling of fibres preventing the penetration of cells. Kaltostat, an alginate-based wound dressing, showed the higher level of swelling and supporte the adhesion of inflammatory cells with limited activation. Knitted dressings showed a limited adhesion of inflammatory cells. In conclusion, this work offers insights about the interactions of these wound dressings with inflammatory cells upon exudate changes thus providing further criteria of choice to clinicians