25507 research outputs found
Sort by
The influence of affective AI literacy on student satisfaction in higher education
Full text linkThis study explores the impact of affective AI literacy on student satisfaction in Pakistan’s evolving higher education sector, which is placing greater emphasis on sustainable education and market-relevant skills. Technology Acceptance Model together with the Cognitive-Affective Theory of Learning with Media (CATLM) is used as theoretical lens for analyzing this investigation. Conducted across three geographically distinct campuses of COMSATS University Islamabad, the research uses a convenience sampling approach. 237 computer science undergraduates participated through an online survey. Measurement items are adapted from established research to ensure validity and reliability, and the data is analyzed using Structural Equation Modeling (SEM). Results indicate that affective AI literacy positively impacts students’ perceptions of AI tools’ usefulness (β = 0.655, p < .001), ease of use (β = 0.613, p < .001), and satisfaction (β = 0.148, p < .01). Perceived usefulness and perceived ease of use are found to mediate student satisfaction, enhancing student engagement and personalization in learning. The study urges higher education to include emotional, ethical, and user-friendly AI considerations into curricula, examining how feelings and attitudes shape students’ perceptions of AI’s usefulness, ease of use, and satisfaction to foster holistic AI literacy. However, limitations include the use of convenience sampling, which focused exclusively on computer science undergraduates from specific campuses, potentially limiting the generalizability to other disciplines or regions. Additionally, future research could explore additional factors like enjoyment, social influence, and academic performance to gain a broader understanding of AI literacy’s impact on student satisfaction
Third Culture Universities: hybrid sociolinguistic fields, hidden language hierarchies, and Artificial Intelligence in transnational Higher Education
Full text linkThis article critically examines the rise of Third Culture Universities (TCUs), a concept to describe international campuses that emerge from Transnational Education (TNE) partnerships between universities in different countries, aiming to blend diverse educational systems and cultures. Such TCUs, often driven by Western academic expansion, conceal persistent power imbalances and sociolinguistic inequalities. These become more apparent as campuses grow, transform, and translate language, learning, and communication. Consequently, this article seeks to utilise Bourdieu’s concepts of 'capital', 'habitus', and ‘symbolic violence’ to explore how these universities drive English-language curricula and Eurocentric standards. This, then, creates grounds for resistance and contested challenge. Employing a postmodern methodology for analysis, the article draws upon a conceptual analysis integrating sociological and sociolinguistic lenses to suggest that TNE partnerships perpetuate social reproduction, making true ‘internationalisation’ remain elusive. The increasing use of Artificial Intelligence (AI), by international students to navigate these challenges and supplement language skills, offers one way forward, we contend. However, such a practice fails to address structural inequality. We conclude by highlighting, therefore, opportunities for resistance and reflexive change in these ‘hidden fields’ of TNE, whilst reframing the role AI will play in shaping power, and practice in academic partnership
Formulation and characterisation of resveratrol-loaded nanostructured lipid carriers for use in combination with scalp cooling therapy to mitigate chemotherapy-induced follicular cytotoxicity and hair loss
Full text linkHair loss represents a highly traumatic side-effect of chemotherapy treatment, it significantly affects psychological well-being, self-esteem and quality-of-life, with the fear of alopecia causing severe anxiety for cancer patients. While effective in eliminating cancer cells, chemotherapy drugs collaterally damage hair follicles resulting in chemotherapy-induced alopecia (CIA). Scalp cooling is a breakthrough treatment for patients, being the only clinically proven method to prevent CIA, with 50–65 % of patients experiencing low grade alopecia (thus negating use of head covers and/or wigs during treatment).
Our recent biological studies showed that optimal cooling effectively protects cells in human hair follicles from chemotherapy drug-mediated damage, whereas suboptimal cooling is less effective. However, combining cooling with an antioxidant that blocks reactive oxygen species (ROS) restores this protective effect against chemotherapy-induced hair follicle damage.
In this study we focused on encapsulating the antioxidant resveratrol (RV) in nanostructured lipid carriers (NLCs) to optimise follicular targeting as a precursor to scalp cooling. We aimed for a particle size above 200 nm to limit systemic absorption and found that the nanoparticles had the desired properties when formulated with propylene glycol dicaprylate as the liquid lipid. RV-loaded NLCs remained stable at 4 °C for >6 months, with less than 10 % variation in their size, polydispersity index (PDI), and zeta potential (ZP). Transmission electron microscopy (TEM) confirmed formation of Type I NLCs, featuring imperfect crystals that suggest a disordered lattice, facilitating RV's presence as disordered crystals or amorphous clusters within the matrix. Skin deposition studies demonstrated that RV-loaded NLCs reach the follicular reservoir within 6 h, confirming their potential for co-application with scalp cooing for combating CIA
Mindful Drawing through the Body
No full textA resource for Mindful Drawing. This video was presented online on 04/06/2025 to the research group Subjective Feminisms and published on Vimeo on 08/02/2026
Predictive Machine Learning Models for assessing the effects of land use and climate change on food affordability in the UK
Full text linkThis research analyses the complex trilateral relationship, land use pattern, climate change and consumer affordability of food products in the UK based on the data set collected from Food Agriculture and Organisation (FAO) from 1961 to 2022. Though agriculture contributes minimally to the UK’s GDP, it plays a very major role in economic stability and in building resilient and sustainable planet. Artificial intelligence is a critical tool that helps in understanding, forecasting and predicting patterns on the complex multidimensional data. This paper aims to apply AI techniques on the data to understand the patterns and dependencies. Initially, the data extracted from the FAO is analysed to understand the trends and the relationship between the attributes is identified using correlation matrix. Several hypotheses are framed, and classification and prediction machine learning algorithms are applied on them. Trend analysis reveals that a decrease in carbon dioxide emission is caused by expansion in the forest land with a very steady high increase in the cost of buying a healthy diet in the UK. Several machine learning models are applied on land use and climate emissions and the support vector regressor shows the highest performance with an R-squared value of 0.96. Furthermore, classification models are applied to get relation between the high and low forest growth regions where the decision tree and the random forest achieved the highest accuracy of 0.8. This research provides valuable insight into the fact that increasing the agriculture land does not reduce the affordability to buy healthy food. Hence, to economically stabilize, the UK should come up with different policies and measures to provide affordable healthy food to people and not just by increasing the agriculture land it can be achieved
Hyderabad’s egg value chain: investigating potential influences on childhood stunting
Full text linkIn 2024, stunting affected over 150 million children under five worldwide, with India being one of the countries with the highest prevalence. The links between stunting, animal health and the health of the environment make a multi-sectoral and interdisciplinary approach essential to combatting childhood stunting in affected countries. This study examined the egg value chain in Hyderabad, India, investigating its potential influences on childhood stunting. Semi-structured interviews with ten key informants gathered data on the egg value chain’s spatial, logistical, economic, and social dimensions. The interviews were analysed through reflexive thematic analysis. A conceptual framework was developed showcasing the links between food value chains and stunting and used as a lens through which to analyse deductively the qualitative data. Reflexive thematic analysis generated themes encompassing the nutritional quality and food safety of eggs accessed by children, how seasonality affected consumption, and the lack of regulation, which forces actors in the chain to depend on trust. The incorporation of a nutrition-sensitive perspective into the value chain analysis revealed stunting-related opportunities and constraints. These findings offer a basis for dialogue with public and private stakeholders to develop targeted interventions that can support efforts to combat stunting in Hyderabad. Additionally, they underscore the method’s utility in generating a comprehensive understanding of stunting determinants and showcase the potential of value chain interventions in addressing the factors contributing to stunting
Noise, resources, and social factors shape space use and movement patterns in a small Amazonian primate
Full text linkNoisy environments tend to mask sounds and disrupt animal behaviour. One of the most common responses from many animal species to noise is leaving or avoiding the noisiest parts of their ranges. Individuals may alter their movement patterns, either rapidly moving away or changing routes in response to increased noise levels. Here, we tested whether urban-dwelling pied tamarins (Saguinus bicolor): (1) preferentially used quieter areas within their home range, (2) increase travel speed, and (3) alter routes to avoid noise levels. To achieve this, we followed nine pied tamarin groups in five forest fragments within the city of Manaus, Brazil, recording their locations and levels of associated ambient noise levels. We also estimated fruit availability at random points within the group’s home ranges. We found that pied tamarins use the quieter areas of their ranges more intensively, but only if the fruit availability is high. Furthermore, groups did not increase their travel speed or change turning angles in response to noise alone, but travel speed increase in noisier areas closer to the border of their home ranges. Thus, pied tamarins do seem to use the quieter areas of their home ranges, but do not hurry to reach them unless their location lies closer to their home range borders. This may occur because higher noise levels at home-range borders could diminish intergroup communication effectiveness. In general, we found that pied tamarin’s use of space and movement patterns are soundscape-related but mediated by resource availability and social aspects
Pollen and sterol content differentially affect solitary bee development
No full textPollen sterols are essential micronutrients for bees, with roles as membrane components, hormone precursors and for regulating gene expression. It is historically accepted that bees are unable to produce sterols de novo or modify them and use them as they occur in pollen. This may require adaptation by bees to specific plants because pollen sterols vary dramatically across taxa. Here we investigated the effects of pollen with different sterol composition or supplementation on development of solitary red mason bees, Osmia bicornis, comparing bee provisioned pollen, with Castanea sativa pollen containing campesterol a key Osmia sterol, a sterol supplemented C. sativa pollen, a polyfloral pollen from which the key campesterol and cholesterol sterols were almost absent, and a combination of polyfloral and C. sativa pollen. DNA barcoding was used to characterise the Osmia pollen, which comprised at least 38 taxa but was mostly Quercus spp. Pollen diet significantly affected the larval final weight, larval development time and development over time. Larvae fed chestnut pollen supplemented with sterols developed significantly faster over time compared to those from the other treatments, including the chestnut pollen with no sterol supplementation. Despite sterol composition being distinctive among pollen diets, sterol profiles were more similar among bees from different pollen foods (i.e., different treatments) than between bees and their pollen foods. Moreover, bees fed on polyfloral pollen had much higher relative concentrations of campesterol (~8.5%) and cholesterol (~26%) than the pollen itself (~3% and ~0.7%, respectively), addressing the possibility that Osmia bees are highly efficient in sterol selective accumulation or possible transformation of sterols in Osmia bees. Our results suggest greater flexibility in sterol handling in bees than previously assumed
Towards Web3- and metaverse-enabled decentralisation of electric vehicle battery closed-loop supply chains: insights from advanced text mining techniques
Full text linkThe rapid transition to electric vehicles has intensified challenges in electric vehicle battery (EVB) closed-loop supply chains (CLSC), particularly regarding material traceability, supply chain transparency, and recycling efficiency. While decentralised technologies, particularly Web3 and Metaverse, offer promising solutions, their integration into EVB CLSC remains fragmented and insufficiently examined. We introduce an Operational Decentralisation Framework enabling a systematic analysis of centralised operations and a critical evaluation of decentralised alternatives as transformational forces. By adopting a holistic perspective, the framework equips firms with strategic guidance for transitioning from centralised structures to decentralised ecosystems. We analyse 588 academic articles and 1,168 industry documents through two advanced text mining techniques – Dynamic Latent Dirichlet Allocation and Burst Detection. Web3 and metaverse can potentially reconfigure the design, manufacturing, end-of-life diagnostics, procurement, waste management, load balancing, capacity planning, inventory management and service operations of two key areas: (1) EVB CLSC operations and (2) EVB circular energy/grid operations. We also found that while blockchain and digital twins show established applications, Web3 and Metaverse applications face significant barriers, including scalability, technology complexity, and expertise gaps, despite their great potentials. Therefore, we propose four visionary models integrating Web3, Metaverse, and AI technologies that have the potential to overcome existing barriers and enable transformative decentralisation. Extending the TOE framework, the study contributes to the theory by developing an integrated framework for evaluating decentralised technology adoption in EVB CLSCs. For practitioners, we provide actionable insights and pathways for technology implementation across different CLSC stages and guidance for addressing key adoption barriers
‘Did I do things right?’ Teachers’ experiences of the policy–practice gap in SEND provision in secondary schools in England
Full text linkStudents with special educational needs and disabilities (SEND) in England are educated in mainstream schools amid rising demand, funding cuts, and staff turnover. While prior research has highlighted teacher workload and lack of training as barriers to inclusion, less attention has been paid to how national policy and accountability structures shape teachers’ perceptions of their capacity to deliver inclusion in practice. This study draws on interviews with 11 mainstream secondary school teachers to examine the systemic and policy-level barriers they face in supporting students with SEND. Thematic analysis revealed three interconnected challenges: the distortive pressures of accountability culture; the mismatch between supply and demand in provision; and teachers’ limited confidence due to insufficient training. Teachers’ accounts illuminate the contradictions between political rhetoric on inclusion and the structures that undermine it. By situating their experiences within the broader context of a shifting policy discourse, this study provides original evidence to suggest that persistent barriers are not simply practical or pedagogical, but symptomatic of long-standing, centrally made choices about funding, training, and accountability. We conclude that proposed reforms to the SEND system provide a generational opportunity to confront these entrenched contradictions, but only if it goes beyond rhetorical commitments to deliver substantive reforms
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇