Research on the role of the physical school environment in areas other than educational psychology has documented the benefits of exposure to nature for cognitive and emotional functioning. Positive effects have been indicated not only after a break in nature in mentally fatigued students but also in students who did not have depleted mental resources when performing cognitive tasks in a green area. We investigated the impact of the physical school environment during a single lesson. We also considered psychophysiological self-regulation as a possible moderator of the relationship between environment and learning. We used data from 101 sixth and seventh graders for self-reported variables. Data about psychophysiological self-regulation, measured as resting cardiac vagal tone, was available for 83 students. In a within-participants research design, we compared the effects of a lesson in the classroom and a lesson in a green urban park close to the school - featuring numerous trees of different sizes, with lush foliage - on participants' affective state, perception of the environmental quality, and conceptual learning. After the lesson in nature, students reported more positive affect and perceived the park as a higher quality environment compared with the indoor classroom. Students with higher cardiac vagal tone learned more in nature, whereas this individual characteristic did not play a role in the classroom environment. Passive exposure to nature during a school lesson has a positive affective impact and may also be beneficial for conceptual learning in combination with a higher ability to flexibly adapt to environmental demands.

Diabetic peripheral neuropathy (DPN) is the leading cause of disturbances in reactive balance control. The repeated, external mechanical perturbations in perturbation-based balance training(PBBT) evoke balance recovery strategies; which subsequently improve reactive balance performance. Using the practice schedule concept of motor learning in the design of PBBT is a relatively new approach related to balance exercises. This study aims to investigate the effects of blocked and random PBBT on reactive balance control and its persistency and transfer to conditions different from those experienced during training. Individuals with DPN will be recruited and randomly allocated to one of the three groups: random, blocked, and control group. Random and blocked PBBT groups will receive single-session balance training, including unexpected perturbations of platform during quiet standing in two directions (anterior and posterior), and three difficulty levels of platform motion (displacement, velocity, and acceleration). Each balance perturbation in blocked group will be repeated over blocks of four trials. For the random group, perturbation sequence will be unpredictable for these four trials in each block. Primary outcomes (i.e., center of pressure variables, reaction time, movement time, and total response time variables) will be assessed at baseline as well as immediately and one day after intervention.

Conceptual problem-based learning model: Promising intervention to enhance prospective science teachers’ critical thinking skills

Orthodontic education has traditionally relied on didactic instruction and rule-based teaching approaches. Increasing emphasis on clinical reasoning and evidence-based treatment planning has led to the integration of conceptual learning strategies-approaches that emphasize understanding relationships between clinical concepts and applying knowledge to diagnostic reasoning, such as case-based and problem-based learning. However, the extent to which these approaches influence diagnostic reasoning and treatment planning skills has not been comprehensively synthesized. To evaluate how conceptual learning strategies are implemented in orthodontic education and examine their reported effects on clinical decision-making, diagnostic accuracy, treatment planning ability, and related educational outcomes. A systematic review was conducted following PRISMA guidelines and registered in PROSPERO (CRD4201025552). Searches of PubMed/MEDLINE, Scopus, Web of Science, Embase, Cochrane Library, and ERIC were performed through January 15, 2026. Eligible studies evaluated conceptual learning approaches in orthodontic education and reported outcomes related to clinical reasoning, treatment planning, knowledge acquisition, or learner engagement. Two reviewers independently screened studies, extracted data, and assessed risk of bias. Twenty-one studies met the inclusion criteria. Interventions included case-based learning, problem-based learning, digital learning platforms, simulation environments, and interactive educational technologies. Across studies, conceptual learning approaches were associated with improvements in diagnostic reasoning, treatment planning ability, knowledge acquisition, and learner engagement, although heterogeneity in study design and outcome measures limited direct comparison. Conceptual learning strategies are increasingly used in orthodontic education and may support development of diagnostic reasoning and treatment planning skills. Further rigorous studies are needed to clarify their long-term educational impact.

Automatically constructing Deep Neural Networks (DNNs) has become a key focus in artificial intelligence research, as their performance is highly dependent on the architecture and parameters of the network. Careful selection of the architecture and parameters for specific tasks can significantly improve the network's output. This study proposes an intelligent and efficient framework for Neural Architecture Search (NAS) by integrating Particle Swarm Optimization (PSO), the concept of Fisher Duty Interval (FDI), Transfer Learning (TL), and graph-based architecture representation. Given the high dependency of DNN performance on architecture design, narrowing the search space and guiding it toward promising regions is essential for improving model effectiveness. In the initial phase, the Fisher Information Matrix (FIM) is used to compute the statistical distance FDI between the target task and a set of previously solved tasks. This serves as the basis for TL, allowing the generation of informed initial particles from previously successful architectures. PSO then performs a multi-level search, balancing global and local exploration. In early iterations, candidate architectures are evaluated relative to baseline architectures, while in later stages, comparison shifts dynamically toward the global best particle. Architectures are encoded as Directed Acyclic Graphs (DAGs) with cell-based modules, allowing for heterogeneous and flexible designs where each cell can have a distinct structure. Additionally, weight inheritance and FIM-based performance estimation reduce the need for full training during each iteration, minimizing computational cost. Overall, this framework leverages prior knowledge, reduces search complexity, and efficiently explores the architecture space to discover models that outperform baseline designs and random search results. The effectiveness of this approach has been tested through classification tasks on well-known datasets, with comprehensive results reported accordingly.

ABSTRACT There is a growing interest in transformative learning in higher education as it provides a unique lens for understanding how to prepare students to think creatively and act purposefully in shaping a more sustainable and equitable world. This study explores student experiences and perceptions of transformative learning within a four-year transdisciplinary combined degree in an Australian university. Drawing on Mezirow’s, Grund et al.’s, and Hoggan’s transformative learning frameworks, we analyse students’ written reflections and semi-structured interviews on their experiences of significant cognitive, emotional, relational, and worldview changes experienced during the course of their four-year study. Students reported a range of transformative experiences, including shifts in the formation of their identity, changes to frames of reference, enriched relationships and enhanced capabilities. The study contributes to higher education theory and practice in three key ways. First, it synthesises transformative learning frameworks that are most appropriate for higher education contexts. Second, it offers a conceptual framework for understanding the kinds of transformative learning experiences that support the development of purposeful careers and meaningful lives. Finally, it calls for stronger links between transformative learning and transdisciplinary concepts and practices in higher education.

ABSTRACT Background With the growing global interest in the aerospace sector, there is an increasing emphasis on strengthening space science education at the elementary level. Innovative and engaging instructional approaches, such as augmented reality (AR), offer new opportunities to support students’ understanding of complex scientific concepts. Purpose This study aims to examine the effectiveness of AR-enhanced digital picture books compared to traditional print picture books in supporting elementary students’ learning of space science concepts. Design and methods A quasi-experimental design with a pre-test – post-test approach was employed. Students were assigned to either an experimental group using AR-enhanced digital picture books or a control group using traditional print picture books. Learning performance, engagement (flow experience), and motivation (technology acceptance) were measured through tests and validated questionnaires. Sample The study involved 50 sixth-grade students in Taiwan. Results The findings indicate that both formats supported students’ learning. However, students using AR-enhanced digital picture books demonstrated higher levels of motivation and deeper conceptual understanding compared to those using traditional print picture books. Conclusions The results suggest that integrating immersive technologies such as AR into educational materials can enhance students’ motivation and conceptual learning in space science. This study provides insights into the design of effective instructional materials and contributes to the integration of educational technology in science education.

Human word concept learning leverages prior knowledge to generalize from limited exemplars, yet its neural implementation remains unclear. We developed a Neural Bayesian Model (NBM) that incorporates neural representational priors to explain word concept learning. Using fMRI, we first constructed a neural prior space from activity elicited by familiar objects (with novel shapes as controls), and then examined neural responses during novel word concept learning based on these objects. The NBM integrating priors from ventral occipitotemporal cortex predicted both neural representations and behavioral generalization, outperforming control models lacking neural priors. In contrast, hippocampal activity supported learning for novel shapes without benefit from the NBM, consistent with prior-free associative mechanisms. Large language models showed weaker alignment with human generalization patterns. These findings dissociate prior-based cortical inference from hippocampal exemplar-associative learning, providing a neural instantiation of Bayesian concept acquisition and clarifying the interplay between semantic and episodic memory systems.

Convergence of chemical and biological space into a unified decision space requires coordinated interactions across the triad- molecular representations, learning algorithms, and explainability. This study establishes a benchmarking framework for Virtual Screening (VS) that integrates this triad to enable consistent, predictive, and interpretable modelling of complex immune-related bioassays. The study employed a wide range of Machine Learning (ML) and Deep Learning (DL) models across three molecular input modalities-descriptors, images, and graphs-while prioritizing predictive reliability and model interpretability. Among classical ML models, Support Vector Machines (SVM) demonstrated the strongest overall performance, while the AttentiveFP architecture outperformed other DL models. Beyond identifying context-specific optimal VS networks, this study derives latent learned concepts that support subsequent hypothesis generation and experimental validation of the interaction between immune targets and ligands.Scientific contributionThis study establishes a benchmarking framework for virtual screening of immune targets by curating representative bioassay datasets, systematically evaluating multiple molecular representations and learning algorithms, and integrating explainability into the best-performing models. By combining Concept Whitening with the AttentiveFP architecture, we introduce an integrated explainable DL framework that maintains predictive reliability while enabling alignment and separation of domain-specific concepts for immune-target virtual screening. In addition, the study highlights the potential for decision- and logical-level model ensembling, identifies target-specific concept relevance in decision-making, and demonstrates the relationship between data characteristics and model behaviour.

The rapid evolution of machine learning techniques, combined with the growing availability of large and diverse data sets, is poised to transform heart failure research and clinical care. This review first provides an overview of key machine learning and artificial intelligence concepts used in heart failure research and then examines how diverse data modalities-including electronic health records, patient registries, biobanks, imaging, telemonitoring, and synthetic data-are leveraged to develop machine learning applications for heart failure diagnosis, prognosis, risk stratification, and personalized treatment strategies. While the potential is considerable, we highlight key barriers to clinical translation, such as data heterogeneity, algorithmic bias, lack of interoperability, and privacy concerns. The review also examines the need for explainable and equitable artificial intelligence systems and evaluates emerging solutions, including Federated Learning and synthetic data generation to address fairness and data privacy challenges. Beyond technical innovations, we underscore the importance of human-centered design, stakeholder engagement, and regulatory readiness. We conclude by identifying future priorities and calling for interdisciplinary collaboration to ensure the scalable, ethical, and effective integration of AI in heart failure management.

Artificial intelligence is creating a new era in university education, impacting both teaching and learning, as well as new research and institutional administration. The current study is a narrative review of the existing literature on the application of artificial intelligence in the future of higher education. Studies published from 2018 to 2025 were identified from various academic search engines using keywords pertaining to the concepts of AI, higher education, adaptive learning, and educational technology. Selected literature was critically analyzed to obtain the major trends, opportunities, and challenges for the integration of artificial intelligence in the University. From the review results, it was revealed that AI technologies can be leveraged for personalized learning, increase student engagement, aid in academic research, streamline administrative tasks, and boost AI-driven decision-making within higher education institutions. The study also raises a number of concerns related to the use of artificial intelligence, such as privacy and bias issues with data, academic integrity, digital inequalities, and the positions of educators in AI-supported learning settings. The paper underscores the need for institutional policies and responsible implementation practices, as well as ethical governance in AI utilization in universities to guarantee transparency and accountability. The study concludes that the future of university education is likely to depend on the collaborative interaction between the educator and intelligent technologies that will create more flexible, inclusive and innovative learning environment.

Machine translation has increasingly shifted toward Neural Machine Translation (NMT) because of its ability to handle input and output sequences of varying lengths. The incorporation of attention mechanisms in NMT systems enables the model to focus on the most relevant parts of the source sentence, rather than relying solely on a fixed representation of the entire input. While NMT improves translation quality by addressing long-range dependencies and contextual understanding, it also requires a large parallel corpus for training, which is a challenge for languages with less resources. The main focus of this research is to give solution for the unique challenges of translating Ayurvedic texts using NMT. Ayurvedic texts have collection of special and scientific words related to medicines and treatments. This makes the translation process more complex and needs very efficient approach for accurate translations. Also, the content of ayurvedic text books is in the form shlokas which is formed using very complex and compound words. In order to simplify the translation process efficiently this work uses a sandhi splitter module and an Anvaya Generator/ word reordering module. In order to develop NMT system for low resource language pair Sanskrit-Malayalam, there is a need of developing a parallel corpus especially for Ayurvedic text books. Also, as the NMT model is proposed for translation it requires a minimum amount of parallel data in the corpus. So, a number of general domain Sanskrit text books with verses, called shlokas, were also considered for developing parallel corpora. The authors developed a parallel corpus for Anvaya Generator, sandhi splitter and translation. Mainly four NMT models were developed trained and tested especially for shlokas as input. The two models are basic transformer model with attention and an encoder-decoder model using Long-Short term Memory (LSTM) with attention. The other two are developed by adding two modules called Sandhi Splitter and Anvaya Generator in the pre-processing stages of the earlier models- Transformer based model and LSTM based model. The limitations of low resources and richness in grammatical structure of Sanskrit- Malayalam language pair are overcome by the concepts of deep learning and the additional modules used in preprocessing stages for developing the models. The models were tested with and without sandhi splitter and Anvaya Generator modules. The transformer-based model integrated with sandhi splitter and Anvaya Generator system achieved a higher average BLEU score of 73.11 and a uni-gram BLEU score of 76.93 for Sanskrit verses to Malayalam translation.

The development of students’ scientific attitudes remains a major concern in physics education, as classroom instruction still tends to prioritize cognitive achievement over cultivating attitudes such as curiosity, respect for evidence, critical reflection, flexibility in thinking, and sensitivity to the environment. This study aimed to examine the effect of integrating the STEM approach with the Predict–Observe–Explain (POE) learning model, supported by the Web S.id platform, on senior high school students’ scientific attitudes when studying static fluids. The study employed a quantitative quasi-experimental method using a nonequivalent control group pretest–posttest design. The participants were 63 eleventh-grade students selected through purposive sampling and divided into an experimental group (32 students) and a control group (31 students). Students’ scientific attitudes were measured using a 12-item questionnaire with acceptable reliability (Cronbach’s alpha = 0.72). The data were analyzed for normality and homogeneity, and an independent-samples t-test and N-gain analysis were conducted. The results showed that the experimental group achieved greater improvement than the control group, with posttest means of 49.16 and 46.61, respectively, and a statistically significant difference between groups (p = 0.032). The effect size was moderate (Cohen’s d = 0.55). N-gain analysis further indicated that all measured indicators of scientific attitude improved more strongly in the experimental group, with the highest gain found in sensitivity in investigating the environment (g = 0.350). The novelty of this study lies in integrating STEM and POE through the Web S.id digital platform, while positioning scientific attitude as the primary outcome in physics learning, particularly in a simple water-dispenser project on static fluids. In conclusion, the STEM-integrated POE learning model supported by Web S.id was effective in fostering students’ scientific attitudes. This study contributes to physics education by providing an empirically supported instructional alternative that integrates project-based STEM learning, inquiry-oriented pedagogy, and digital media to strengthen affective outcomes alongside conceptual learning.

Midwifery continuity of care (MCoC) is a key aspect of woman-centred midwifery care, associated with improved outcomes for both women and midwives. In many countries, maternity care is fragmented, and MCoC uncommon in clinical practice and midwifery education. To address this gap, a pedagogical project entitled "Midwifery Student All the Way" was introduced in a midwifery programme in southern Sweden. The project was designed to provide students with MCoC experience throughout pregnancy, birth, and postpartum care. This study aimed to describe midwifery students' conceptions of learning within a pedagogical midwifery continuity of care project. A qualitative study with a phenomenographic approach was conducted to explore midwifery students' conceptions of the skills they acquired through participation in the project. Focus group interviews were conducted with a total of 22 students at the end of their education. Data were analysed using the seven-step phenomenographic method. Five categories and thirteen conceptions were identified. The result showed that students developed knowledge in a unique way through relational learning and by being with the women. The students needed to navigate challenges and create their own role in the project. They benefited from a collaborative experience with their peers contributing to their learning. The midwifery continuity of care experience project was proven beneficial for student-centred learning, even within a fragmented context of clinical healthcare placement and in a developing phase. Students described participation in the MCoCE project, which offered continuous engagement with women during childbearing, as highly enriching their learning through active involvement, thus boosting and strengthening their professional confidence and identity. They described a possibility to develop meaningful relationships, internalise the core principles of midwifery, and acquire essential clinical competencies and perspectives that often are less accessible through ordinary education. Identified challenges must be addressed when formalising the project.

Visual relational concepts-defined by patterns of relationships between entities-are thought to require structured, compositional representations with explicit role information about each entity. Analogical mapping over compositional representations is a key strategy for acquiring such concepts, but in complex situations with many entities and relations, this process can be cognitively demanding. As a result, learning may occur over feature-based representations, where exemplars are encoded as unstructured lists of entities and relations, losing crucial role information and limiting generalizability. To reduce the cognitive load of analogical mapping, we explored the effectiveness of two visuospatial training aids: (a) spatially organizing exemplars by category to facilitate comparisons and (b) using color coding to highlight the roles of entities within each exemplar. Across three experiments, we examined whether these visuospatial aids improve learning rates on the Synthetic Visual Reasoning Test (SVRT), a collection of 23 problems that require learning relational concepts. Our results showed that displays of previous instances that spatially sorted them into positive and negative sets led to faster concept learning. Learning was faster overall when problems were ordered easy-to-hard rather than randomly, but sorted displays were more effective in either case. Color coding proved beneficial only when colors unambiguously and nonredundantly linked entities that played corresponding roles; when color coding did not support a clear mapping, it interfered with learning. These findings suggest that rapid learning of relational concepts can be facilitated by display characteristics that support analogical mapping by comparisons. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

This work describes the use of participatory action research to develop an artificial intelligence (AI)-augmented, peer-driven, case-based, and simulation-enhanced framework for senior emergency medicine trainees. It has been applied to enhance knowledge acquisition for small-group self-directed study in resuscitation medicine. Trainees engaged in structured learning cycles over 6months, based on the principles of 'desirable-difficulty' and deliberate-practice. It incorporated peer-selected pre-reading, case-based discussions, high-fidelity simulations, and spaced-repetition flashcard review. A key innovation is the use of generative AI tools to supplement these activities, and follow evidence-based prompt engineering. The participants refined self-study methods through iterative evaluation. AI-generated questions facilitated retrieval-based learning, and flashcard integration enhanced knowledge retention. Simulation-based reinforcement contributed to the 'desirable-difficulty' through the clinical application of learned concepts. Self-reported recall improved over time. This structured, self-directed approach supports effective learning in resuscitation medicine. AI and peer-driven strategies augment knowledge retention. This methodology offers adaptability for broader medical education settings.

Gamification has gained momentum in STEM education as a way to boost student engagement, motivation, and conceptual learning. A wide variety of games, from short in-class activities to long-format student-built projects, are used across disciplines. However, few studies have examined why different game formats succeed or fail across varying contexts. In this perspective piece, we examine how games have been used in higher education STEM classrooms and highlight key contrasts between game types, implementation goals, and learning outcomes. Building from this foundation, we explore the added value of combining game-based learning (GBL) with design-based learning (DBL), particularly through student-led game design. We suggest that student-led game design, which incorporates both GBL and DBL principles, provides instructors with a flexible way to align games with course content, promote systems thinking, and encourage collaboration. Student-led game design is also highly adaptable to online learning environments, offering a way to enhance community and communication, which are typically challenged in this instructional format. Overall, we found that gamification in STEM is most effective when instructors consider both the diversity of game structures and how these support specific learning outcomes. Student-led game design is a flexible, underused strategy that can engage and motivate college-level students across scientific disciplines.

This study aims to analyze the mathematical values contained in the traditional house of the Samin community and determine the application to use the concepts in mathematics learning. Ethnographic approach and triangulasi method are used in qualitative research analysis. Data collection techniques were carried out by literature study, participatory observation, documentation and interviews. Triangulasi is used for data variation and validation while ethnografi is used to study and develop cultural understanding. The results showed that there are ednomathematics concepts that are relevant to be applied to mathematics learning such as the shape of the roof, poles, doors, windows and the middle room. The discovery of the concept of flat shapes (square, rectangle, triangle, trapezoid), spatial shapes (block, prism), geometrict transformations (translation, rotation, reflection), and similarity. The mathematical concept in this research can be applied to the school curriculum from elementary to high school. Practical value, this research help students understand mathematics through their own enviroment and cultural. Cultural value, ednomathematics based learning strengthens local identity and cultural presenvation. Use value, the math learning approach is more contextual, fun and relevant ro real life.

Training medical vision-language models (VLMs) typically demands millions of image-text pairs to achieve versatility and reasoning, posing significant challenges in data acquisition. We propose ConceptVLM, a novel data-efficient fine-tuning paradigm that transforms general-domain VLMs into specialized medical ones with minimal labeled data, integrating medical knowledge without disrupting the model's existing general capabilities. Central to our approach is a key concept-aware training strategy, building a structured medical concept dictionary and employing masked attention to guide the model's focus toward essential clinical concepts. This focused fine-tuning enhances domain-specific comprehension while preserving the model's reasoning abilities and response diversity. Experiments across multimodal medical benchmarks show ConceptVLM achieves state-of-the-art results using only 1% of the original training data, outperforming traditional methods reliant on large-scale QA datasets. These findings challenge the prevailing reliance on extensive annotated corpora, demonstrating key concept-guided tuning as a viable path to developing cognitively capable medical VLMs.

This study examines the educational effects of an argumentation-supported model-based learning (AB-MBL) approach on gifted sixth-grade students’ conceptual learning in biodiversity and sustainable living. The research was conducted with 21 gifted students attending a Science and Art Center (SAC) in Türkiye. During the four-week process, students engaged in modeling and argumentation activities designed to support learning in biodiversity and sustainable living education. Quantitative data were collected using a two-tier Biodiversity Conceptual Understanding Test administered as a pre-test and post-test, while qualitative data were obtained from students’ learning journals. Quantitative findings revealed a statistically significant increase in gifted students’ conceptual understanding of biodiversity after the intervention. Analysis of the learning journals indicated increased awareness of biodiversity conservation and active engagement in the learning process. The findings suggest that AB-MBL was associated with improvements in gifted students’ conceptual understanding and shows promise as a pedagogical approach for biodiversity and sustainable living education.