Mesh-free numerical methods provide flexible discretisations for complex geometries; however, classical meshless discrete differential operators typically trade low computational cost for limited accuracy or high accuracy for substantial per-stencil computation. We introduce a parametrised framework for learning mesh-free discrete differential operators using a graph neural network trained via polynomial moment constraints derived from truncated Taylor expansions. The model maps local stencils relative positions directly to discrete operator weights. The current work demonstrates that neural networks can learn classical polynomial consistency while retaining robustness to irregular neighbourhood geometry. The learned operators depend only on local geometry, are resolution-agnostic, and can be reused across particle configurations and governing equations. We evaluate the framework using standard numerical analysis diagnostics, showing improved accuracy over Smoothed Particle Hydrodynamics, and a favourable accuracy-cost trade-off relative to a representative high-order consistent mesh-free method in the moderate-accuracy regime. Applicability is demonstrated by solving the weakly compressible Navier-Stokes equations using the learned operators.
Computational models of memory have achieved considerable success by formalizing how traces are encoded, cued, and retrieved. However, the role that individual linguistic experience plays in shaping representational structure has been relatively understudied. Where the issue has been examined, most models assume that a common population-level semantic space suffices, effectively treating individual variation in language experience as noise rather than signal. We test the theoretical adequacy of this assumption. In a large-scale experiment (N = 478), participants completed a cued recall task in which identical target words were paired with cues drawn from eight different genre-specific semantic spaces (mystery, fantasy, horror, literary fiction, romance, science fiction, thriller, and non-fiction). Participants reported their reading habits across the eight genres, and personalized semantic representations were constructed by weighting genre-specific corpora proportionally to each participant's reading profile. Two complementary analyses were conducted. In a model-free analysis, cosine similarity computed within each participant's personalized semantic space predicted recall and omission outcomes more reliably than similarity derived from generic semantic spaces. In a computational analysis, substituting personalized representations into the embedded Computational Framework of Memory improved cue–target-level predictions over generic alternatives, accounting for approximately 4 percentage points more explained variance. These findings provide proof of concept that variation in individual linguistic experience shapes semantic structure in ways that are both behaviourally detectable and computationally tractable. More broadly, they suggest that the predictive limits of memory models may reside not only in their retrieval mechanisms but also in the fidelity of the representations they assume.
Understanding safety in complex socio-technical systems requires analytical approaches that move beyond linear accident models to examine how interactions across organisational, technical and operational elements shape safety outcomes. This study applies two systemic safety analysis approaches, Causal Analysis based on Systems Theory (CAST) and the Functional Resonance Analysis Method (FRAM), using well-documented aviation investigation data. The study examines how different systemic methods model system behaviour, frame causality and performance variability, and generate different forms of safety recommendations. CAST identified cross-level control and feedback weaknesses that enabled the wrong-surface alignment under the runway-closure night configuration. FRAM showed how coupled performance variability, including missed runway-closure cueing and ambiguous visual cues, shaped the development of misalignment risk while also clarifying the recovery pathway that enabled the go-around. This study suggests that CAST and FRAM are best used as complementary lenses for systemic event analysis. CAST supports governance and assurance redesign, while FRAM informs operational guardrails and variability management under uncertainty.
Self-concept clarity, the degree to which an individual has a well-defined and stable sense of self, is a well-documented factor in mental health conditions, particularly eating disorders. Difficulties with self-concept clarity are also reported among gender diverse and neurodivergent people, who are overrepresented in eating disorder populations. This cross-sectional study examined associations between self-concept clarity (Self-Concept Clarity Scale), autistic traits (Autism Spectrum Quotient), ADHD traits (Adult ADHD Self-Report Scale), gender diversity (Gender Self-Report), and disordered eating, a pattern of atypical eating behaviors and attitudes including food restriction and binge eating (Eating Disorder Examination Questionnaire). Gender diversity was assessed as binary (identity opposite to sex assigned at birth) and nonbinary traits (identity neither female nor male). Participants were 492 UK adults (324 assigned female at birth; 98.6
Glioblastoma multiforme is characterised by resistance to conventional treatments via sustained pro-inflammatory signalling. This study investigated whether secretomes and small extracellular vesicles (sEVs) from human oral mucosa neural crest-derived stem cells (NCSCs) could disrupt multiple glioblastoma features by rebalancing pro- and anti-inflammatory pathways. NCSCs were characterised, and their secretomes/sEVs isolated, and analysed using a range of state-of-the-art methods. Anti-tumour effects were evaluated using multiple glioblastoma cell lines (U251, U373, U87) through viability, proliferation, migration, and tumourigenicity assays. Mechanistic studies employed dual nuclear factor-κB (NF-κB) and quadruple NF-κB/interferon regulatory factor 3 (IRF3) reporter systems, cytokine arrays, and immunocytochemistry. Therapeutic potential was assessed using temozolomide (TMZ) chemosensitivity assays and organotypic mouse brain slice models. NCSC secretomes and sEVs demonstrated consistent anti-glioblastoma activity across all functional assays. Both products potently suppressed NF-κB activation across multiple pro-inflammatory stimuli whilst simultaneously enhancing IRF3 nuclear translocation and transcriptional activity. This dual pathway modulation reprogrammed glioblastoma cytokine secretion towards anti-inflammatory profiles, and inhibited tumourigenicity in both 3D culture and ex vivo brain slice models. Notably, secretomes and sEVs enhanced the efficacy of TMZ, reducing colony size compared to monotherapy, without compromising tissue viability. These findings demonstrate the first evidence of dual NF-κB/IRF3 pathway rebalancing by NCSC-derived products in glioblastoma. The simultaneous suppression of tumour-promoting inflammation whilst enhancing anti-tumour immune signalling represents a novel therapeutic paradigm that addresses multiple resistance mechanisms simultaneously. This strategy redefines stem cell-based therapy for glioblastoma by integrating immunomodulation with chemosensitisation in a single, Good Manufacturing Practice (GMP)-compatible product.
