Search Results (684)

Spatial updating—the ability to continuously revise spatial representations during locomotion—is fundamental to adaptive navigation and depends on flexible reference frames. Although previous research has established independent effects of field cognitive style and scene layout on spatial performance, their interaction and underlying neural substrates remain poorly understood. The present study examined how field dependence–independence and environment geometry jointly modulate spatial updating by combing the judgment of relative direction (JRD) paradigm with functional near-infrared spectroscopy (fNIRS). Forty participants were recruited and assigned to two groups (20 field-independent [FI] and 20 field-dependent [FD]) based on Embedded Figures Test scores. They completed directional pointing tasks in two virtual environments: a geometrically structured rectangular spaces affording explicit orthogonal reference axes and an ambiguous oval environments devoid of stable global geometric anchors. Behaviorally, FI individuals exhibited shorter response time in rectangular layouts yet superior accuracy in oval layouts relative to FD individuals. Neurally, the middle frontal gyrus (MFG) emerged as a critical locus exhibiting a significant interaction effect between cognitive style and environmental layout. Significant main effects of field cognitive style were observed in the precentral gyrus, superior parietal lobule, and paracentral lobule, with FI individuals showing greater oxyhemoglobin (HbO) elevation than FD participants. Collectively, these findings may tentatively suggest an interpretation that FI individuals flexibly alternate between internal egocentric and external allocentric reference frames during spatial information processing, whereas FD individuals predominantly rely on inherent structural cues embedded in the external environment. These findings may reflect cortical hemodynamic correlates of field cognitive style differences during spatial processing, and may offer empirical references for relevant cognitive neuroscience research and subsequent exploratory applications. Full article

Cell migration is a fundamental biological process regulated by interactions between cells and extracellular matrix. Although topographical cues are known to influence cell behaviors, directional migration across three-dimensional (3D) sloped steps remains poorly understood. Here, 3D sloped steps with patterned TiO_x surfaces were fabricated to investigate topography-guided cell migration in complex 3D microenvironments. The ultrathin TiO_x layers were patterned along the bottom, sidewall, and top regions of the steps, providing continuous guidance during cell migration up or down the steps. MC3T3-E1 cells were confined to the patterned regions and exhibited contact-guided migration along the asymmetrical arrowhead patterns. Forward and reverse arrowheads were introduced to evaluate the effect of geometrical asymmetry on cell migration directionality. Forward arrowheads preferentially guided cells from the bottom to the top of steps, whereas reverse arrowheads promoted migration down the steps, demonstrating reversible control of cell migration direction through arrowhead orientation. Analysis of cell morphology revealed that ultrathin TiO_x topographies influenced lamellipodia orientation and cell adhesion, providing mechanistic insights into geometry-mediated control of cell migration direction. These findings demonstrate that guiding pattern asymmetry can be used to regulate the speed and directionality of cell migration across sloped steps, which can be applied to control cell migration behaviors on engineered 3D platforms. Full article

Autonomous obstacle avoidance for UAVs in dynamic obstacle-dominated complex environments must address time-varying local collision risks from multiple directions under the constraints imposed by local sensing, environmental uncertainty, execution safety, and limited onboard computation. Planning-based methods often require frequent replanning or explicit obstacle prediction, whereas conventional reinforcement learning policies may produce myopic decisions under partial observability. To address these limitations, this study proposes a dynamic obstacle-avoidance framework that combines a temporal LiDAR representation with safety-aware action correction in recurrent reinforcement learning. Multi-layer LiDAR observations are constructed using sector-wise minimum pooling. Adjacent two-frame stacking and a CNN-LSTM architecture are then used to extract local geometric structures and short-term dynamic cues, and a velocity-control policy is optimized using Recurrent PPO. In addition, a smooth velocity-projection safety shield is introduced to modify policy outputs and reduce collision risk during both training and policy execution. Experiments conducted in mixed static–dynamic obstacle scenarios based on Gym-PyBullet-Drones show that the proposed method achieves an average success rate of 91.9% across four test configurations, with an average online computation time of 0.78 ms. Ablation studies further support the contributions of two-frame observations, LSTM-based temporal modeling, and the safety shield. Full article

Soccer-video analysis centers on pitch-plane tracking, but camera-view depth cues such as occlusion and goal-area structure are not fully represented on the field plane. Synthetic benchmarks provide dense supervision unavailable for real broadcasts, but whether adaptation yields predictions that are reproducible across matches and operationally feasible remains unclear. We evaluate a Depth Anything V2 model adapted to SoccerNet-Depth with four components: Unaligned MDE accuracy, scale-and-shift aligned diagnostic, match-to-match reliability, and accuracy–cost trade-off. The model achieves an unaligned validation AbsRel of 0.00372. The aligned diagnostic shows that Base DAv2 retained substantial scene-depth structure, whereas SoccerNet adaptation enabled direct compatibility with the normalized target without per-frame ground-truth fitting. Relative to the VKITTI-fine-tuned reference, the adaptation improved all eight metrics in all 21 validation matches, with paired Wilcoxon tests significant after Bonferroni correction. On the challenge split, it reduced AbsRel by 34.1% versus the official baseline. The higher-resolution configuration improved the validation AbsRel by 5.9%, while the default retained a better accuracy–cost balance. At 401.57 ms per frame, the default is suited to post-match analysis, not live or near-real-time use. The study contributes a benchmark-scoped adaptation case study and protocol for foundation MDE on SoccerNet-Depth. Full article

Generation Z consumers are reshaping food consumption patterns in urban digital environments, particularly in restaurant contexts characterized by high choice complexity and uncertainty. In Mexico, the evolution of the restaurant industry has intensified the need to understand how digital cues shape consumer food choices. Addressing this gap, this study examines how Social Media Marketing (SMM), Social Media electronic Word of Mouth (Social Media eWOM), and Social Media Influencers (SMIs) shape food consumption intention among Generation Z in Mexico City. Grounded in the Stimulus–Organism–Response (SOR) model and integrating the attitudinal foundations of the Theory of Reasoned Action (TRA) and the Theory of Planned Behavior (TPB), this study analyzes how these digital factors impact food consumption intention (operationalized as restaurant purchase intention) through the mediating psychological mechanism of Consumer Brand Engagement (CBE). A quantitative, non-experimental design was employed using a sample of 406 respondents, and data were analyzed through Structural Equation Modeling (SEM). The results indicate that the model explains 73.6% of the variance in food consumption intention. SMM emerged as the strongest direct predictor, followed by Social Media eWOM and SMIs. Crucially, CBE mediates only the relationship between influencers and consumption intention. Conversely, both SMM and Social Media eWOM exert direct effects that bypass affective engagement. These findings highlight the role of digital ecosystems as cognitive proxies in restaurant selection, providing actionable insights for restaurant SMEs to optimize digital strategies and enhance economic resilience. They also suggest potential implications for healthier and more sustainable urban food environments. Full article

When people recount personal memories, they often refer to people, places, and events indirectly, relying on contextual cues rather than explicit names. Such implicit references are central to reminiscence narratives: first-person accounts of lived experience used in therapeutic, archival, and social settings. They pose a difficult computational problem because the intended entity must be inferred from dispersed narrative evidence rather than from a local mention. We introduce IRC-Bench, the Implicit Reminiscence Context Benchmark, for evaluating implicit entity recognition in reminiscence transcripts. The benchmark targets non-locality: entity-identifying cues are distributed across multiple, non-contiguous clauses, unlike named entity recognition, entity linking, or coreference resolution. IRC-Bench comprises 25,136 samples constructed from 12,337 Wikidata-linked entities across 1994 transcripts spanning 11 thematic domains. Each sample pairs an Entity-Grounded Narrative, in which the target entity is explicitly mentioned, with an Entity-Elided Narrative, in which direct mentions are removed. We evaluate 19 configurations across LLM generation, dense retrieval, RAG, and fine-tuning. QLoRA-adapted Llama 3.1 8B performs best in the open-world setting (38.94% exact match; 51.59% Jaccard), while fine-tuned DPR leads in closed-world retrieval (35.38% Hit@1; 71.49% Hit@10). We release IRC-Bench with data, code, and evaluation tools. Full article

Cancer remains one of the most complex diseases to study and treat, with tumour microenvironment heterogeneity and therapeutic resistance continuing to limit clinical progress. Biomaterials-based nanoparticles have emerged as versatile platforms that not only advance understanding of cancer biology but also enable innovative therapeutic strategies. As mechanistic tools, nanoparticles can be used to investigate extracellular matrix interactions, mechanotransduction pathways, drug resistance, and tumour–immune crosstalk, providing insights into how physical and biochemical cues influence disease progression. Therapeutically, engineered nanoparticle systems have been developed for the targeted delivery of chemotherapeutics, nucleic acids, and immunomodulatory agents, incorporating features such as stimuli-responsive release, multifunctionality, and theranostic capabilities. Recent advances in patient-derived tumour models, high-throughput screening platforms, and artificial intelligence-assisted design are further driving the development of precision nanomedicine. Despite ongoing challenges related to biodistribution, safety, manufacturing scalability, and regulatory approval, nanoparticle-based biomaterials offer significant opportunities to bridge fundamental cancer research and clinical translation. This review highlights recent mechanistic and therapeutic advances, discusses key translational barriers, and outlines future directions at the interface of biomaterials, nanotechnology, and oncology. Full article

Biochar can alter soil physical conditions, microbial carbon processing, and soil organic carbon (SOC) responses under fluctuating moisture, yet how these changes are coordinated remains insufficiently understood. We conducted a two-season greenhouse pot experiment to examine the initial (first-year) and residual (second-year) effects of wheat-straw biochar (WSB) and softwood biochar (SWB) under conventional deficit irrigation (CDI) and alternate drying–wetting cycles (DWC). Compared with unamended soil, biochar amendment improved water-dispersible microaggregate-size distribution, mean microaggregate size, and water-holding capacity, which contributed to reduced soil thermal conductivity (STC) by 6.0–14.2%. Biochar application also improved microbial carbon use efficiency (CUE) by 41.1–52.3%, with WSB generally showing stronger and more persistent effects than SWB. Relative to CDI, DWC increased soil respiration rate by 14.9–48.8% but decreased CUE by 7.4–10.2% and SOC by 3.0–10.3%, indicating a shift toward greater respiratory carbon loss under repeated moisture fluctuations. Biochar amendment increased SOC across both seasons, particularly under WSB, and partially alleviated the DWC-associated reductions in CUE and SOC. Correlation analyses showed that lower STC was associated with higher CUE and SOC, but these relationships should be interpreted as coordinated associations rather than direct evidence of a causal thermal-regulation mechanism. Principal component and random forest analyses further highlighted STC as a prominent variable associated with variation in CUE and SOC among the measured soil attributes. These findings indicate that biochar-mediated changes in soil physical conditions are closely associated with microbial CUE and SOC responses under drying–wetting cycles, wherein soil thermal properties may represent an important physical dimension of these carbon responses. Full article

Background/Objectives: Mild cognitive impairment (MCI) is a clinically important state associated with an increased risk of future cognitive decline and dementia. Behaviour change interventions may support risk reduction and self-management in cognitively vulnerable adults. However, the extent to which the transtheoretical model (TTM) has been used in this population has not been clearly mapped. This scoping review examined TTM-based, TTM-informed, and TTM-congruent behaviour change interventions for adults with MCI, subjective cognitive concerns, or elevated dementia risk. Methods: This scoping review followed Joanna Briggs Institute guidance and was reported in accordance with PRISMA-ScR. The protocol was prospectively registered on the Open Science Framework. PubMed, PsycINFO, ScienceDirect, Scopus, Web of Science, CENTRAL, ProQuest Dissertations, and medRxiv were searched from inception to 30 September 2025. Eligible studies included randomised, nonrandomised, quasi-experimental, and qualitative designs. Data were charted using a piloted extraction form and were synthesised narratively. Results: Eight unique studies, represented across nine publications, were included. Only one trial explicitly operationalised the TTM in a clinically defined MCI cohort; most studies were more appropriately classified as TTM-informed or TTM-congruent. Recurrent intervention components included readiness alignment, goal setting, self-monitoring, personalised feedback, prompts and cues, problem solving, reinforcement, and relational support. Behavioural outcomes were more consistently favourable than cognitive outcomes, particularly for adherence, self-management, diet, and sustained physical activity engagement. Cognitive findings were heterogeneous: some smaller studies reported short-term improvements, whereas the largest rigorous trial found no significant cognitive benefit. Conclusions: Current evidence does not support strong claims regarding TTM-specific cognitive efficacy in MCI. Instead, it suggests that TTM-informed and TTM-congruent interventions may be useful for strengthening behavioural regulation, risk reduction, and maintenance of health-related routines in cognitively vulnerable adults. More rigorous studies are needed to test the TTM constructs prospectively and to determine whether proximal behavioural change translates into durable cognitive or functional benefit. Full article

Eye-movement studies of manual production often average gaze across an entire trial, obscuring how visual information use changes once actions begin. We separated the pre-writing and writing phases in a fixed progressive Chinese calligraphy task. Thirty-seven postgraduate students completed two style-guided transfer (SGT) pages, a worked example, and two evolution-based mapping (EBM) pages; 34 contributed usable gaze data. On SGT pages, reference allocation fell from 0.626 before writing to 0.131 during writing, whereas the share of reference viewing directed to diagnostic tokens rose from 0.473 to 0.601. On EBM pages, allocation to the cue-plus-context display fell from 0.825 to 0.447 after pen onset but remained substantial; cue share and context coverage also declined. Participant-level process blocks did not improve quality models. In exploratory page-level EBM analyses, greater pre-writing context coverage was associated with higher product quality. These findings identify pen onset as a useful boundary for analyzing visual information use in constrained production: external sampling is greatest before writing, and task-specific re-access persists during execution. Because the task order was fixed, page-family differences cannot be separated from practice or scaffolding. Phase-specific area-of-interest measures can therefore add process information to product scores without treating gaze as a direct measure of cognition. Full article

Low-resource hate speech detection remains a challenging task for Indonesian social media due to limited labeled annotations, highly informal linguistic expressions, and substantial lexical variability. Under such conditions, purely supervised transformer models often suffer from unstable semantic generalization, while conventional pseudo-labeling methods are vulnerable to noisy unlabeled sample propagation. To address these limitations, this study proposes a hybrid feature-enhanced IndoBERT framework integrated with a controlled semi-supervised learning strategy. The proposed model combines contextual IndoBERT embeddings with abusive lexicon cues, handcrafted linguistic indicators, and TF-IDF–SVD statistical representations through a lightweight concatenation–projection feature fusion mechanism, while unlabeled data are incorporated via adaptive confidence thresholding and class-balanced pseudo-label selection to improve pseudo-label reliability. Extensive experiments were conducted under realistic low-resource supervision settings using only 5%, 10%, and 20% labeled data, and the proposed framework was systematically compared against representative baselines, including sparse lexical machine learning models, shallow neural architectures, multilingual transformers, IndoBERTweet, naive pseudo-labeling, and LLM-based prompting. The results show that model effectiveness is strongly supervision-dependent. Under the most extreme low-resource setting, compact statistical augmentation provides the most stable complementary signal, whereas under moderate low-resource supervision, the full hybrid representation combined with controlled semi-supervised learning yields the strongest and most consistent gains. The proposed Hybrid IndoBERT + controlled SSL framework outperforms all baselines at the 20% labeled setting, reaching an accuracy of 0.8654, Macro-F1 of 0.8633, and ROC-AUC of 0.9334. Additional analyses of pseudo-label reliability, calibration behavior, computational efficiency, and qualitative error patterns further show that the proposed framework improves low-resource robustness while maintaining comparable inference-time efficiency. These findings demonstrate that low-resource hate speech detection benefits most from the staged integration of contextual semantic modeling, interpretable linguistic cues, global lexical–statistical structure, and carefully regulated unlabeled data exploitation. Additional experiments using GPT-4o-mini and Llama-3.1-8B further demonstrate that the proposed framework remains competitive against general-purpose large language model prompting approaches under low-resource Indonesian hate speech detection scenarios. The proposed framework provides a practical and reproducible direction for hate speech detection in annotation-constrained social media environments. Full article

Stem cells, with their self-renewal and multi-lineage differentiation potential, hold promise for tissue repair and intractable diseases treatment. Yet clinical translation of stem cell therapies has long been hindered by insufficient scalable stem cell manufacturing, stemness loss and functional decline in 2D expansion, and poor post-transplantation cell retention, unregulated fate control. Programmable microcarriers (MCs) paired with 3D dynamic culture offer an emerging strategy to address these bottlenecks and enable stem cell fate regulation. In this review, we systematically review advanced MC fabrication strategies for stem cell fate regulation, comparing features of emerging technologies (microfluidics, electrospraying, in-air microfluidics, integrated in situ functionalization) and their implications for programmable MC control and scalable manufacturing. We analyze how MCs modulate stem cell behaviors (adhesion, proliferation, stemness maintenance, differentiation) via synergistic static physicochemical cues and dynamic stimuli-responsive properties. We map the latest advances in functionalized MC-mediated stem cell therapy across osteochondral defects, autoimmune, skin, ophthalmic and neurodegenerative diseases. Finally, we pinpoint unresolved challenges for clinical translation of MC–stem cell system and outline key future research directions. This review offers a systematic roadmap for advancing programmable MC fabrication, clinical-grade stem cell biomanufacturing, and precise cell therapy development. Full article

This study examines how live-streaming cues influence impulsive purchase intention in fresh-fruit e-commerce, where consumers face substantial quality uncertainty and limited opportunities for pre-purchase inspection. Drawing on the Stimulus–Organism–Response (S-O-R) framework, we examine five stimuli—anchor professionalism, anchor interactivity, visual attractiveness, price discount, and scarcity—and test whether perceived value (cognitive) and positive emotions (affective) operate as parallel mediators. Based on survey data from 353 Chinese consumers, the results show that anchor professionalism, anchor interactivity, price discount, and scarcity are positively associated with impulsive purchase intention both directly and indirectly through perceived value and positive emotions, whereas visual presentation follows a different pattern. Contrary to the common assumption that vividness primarily triggers emotional impulse, visual attractiveness does not exhibit a robust direct effect on purchase intention; instead, its influence is transmitted dominantly through cognitive perceived value rather than affective positive emotions. This finding suggests that, in high-uncertainty perishable categories, vivid presentation is more consequential when it helps consumers evaluate product value than when it merely stimulates affective reactions. The study offers targeted implications for S-O-R theory and provides practical guidance for platform design and promotional disclosure in real-time e-commerce. Full article

UAV remote sensing provides centimetre-level imagery for fine-grained cotton growth monitoring, yet existing segmentation models face three challenges: cotton fields exhibit a pronounced row and column structure that standard convolutions struggle to capture; conventional decoders fuse features statically, suppressing fine boundary cues; and the pixel-level class distribution is severely imbalanced. We present DCA-DeepLab, built on DeepLabv3+ with three task-specific components: a Dual-Coordinate Attention Gating (DCAG) module that decouples horizontal and vertical dependencies to encode row and column structures; a Multi-Scale Attention-Guided Modulated Feature Merging (MSAM-MFM) module that reweights semantic and detail features at each location; and an adaptive pixel-level modulated focal loss (APMFL), which focuses training on hard, minority-class pixels. We construct a cotton growth dataset of 11,745 UAV patches with four semantic classes. On this dataset and the public LoveDA benchmark, DCA-DeepLab attained the highest mIoU among the compared methods (51.74% and 51.71%), exceeding the strongest cotton baseline by 1.10 percentage points. Relative to DeepLabv3+, the Vigorous and Sparse minority-class IoUs improved by 3.51 and 1.91 percentage points, respectively, and Vigorous recall rose from 51.85% to 60.04%, with only 3.9% more parameters. These results show that encoding directional structure and adaptively balancing class contributions benefits fine-grained UAV crop segmentation. Full article

City-scale unmanned aerial vehicle vision-and-language navigation (UAV-VLN) requires accurate upstream target localization from an overhead map, onboard observation, and language description. Existing VLM-based methods often treat road names, landmarks, and spatial relations as raw text, leaving the model to search a large map and implicitly infer geometric constraints. This paper proposes GPOD, an inference-time candidate-prior interface for the upstream target-localization stage in city-scale UAV-VLN. GPOD converts language anchors, spatial relations, target-category cues, static map objects, and vehicle detections into ranked candidate priors through branch-specific candidate generation, thereby reformulating unconstrained full-map coordinate regression as candidate-prior-conditioned coordinate prediction. The static branch aligns language constraints with map-object geometries, while the dynamic branch uses YOLOv8l-VisDrone with Slicing Aided Hyper Inference (SAHI) to construct detection-conditioned vehicle candidates. In the GPOD-VLM setting, ranked candidates are injected as structured spatial prompts and the base VLM predicts the final continuous coordinates; GPOD-Direct is a candidate-direct diagnostic variant that directly uses candidate centers without VLM coordinate regression. On the CityNav localization protocol, GPOD improves FlightGPT Overall SR@20m from 15.23% to 25.61% and consistently reduces Mean Navigation Error (Mean NE) across splits and backbones. On Val-Unseen, GPOD-Direct (Top-1) reaches 32.59% SR@20m, showing that ranked candidate priors provide strong discrete localization signals. These results show that inference-time candidate priors can reduce city-scale search ambiguity without updating the base VLM parameters, while also revealing a candidate-utilization gap in the current prompt-based continuous coordinate-regression interface. Full article