From peripersonal space to cognitive maps: An evolutionary perspective.

Navigating complex social environments requires understanding how individuals are connected within social networks. People are highly efficient at integrating dyadic relationships into larger network structures. However, social networks are embedded within different types of groups (e.g., task groups vs. social categories), raising the question of how group typology shapes the learning and representation of social networks. Across four experiments, we demonstrate that group-related network schemas-specifically, expectations that task groups are more interconnected and centralized than social categories-systematically guide social network learning and representation. Participants exhibited prior expectations about the relational structure of different group types, and memory for social relationships was enhanced when the learned network structure aligned with these schematic expectations, producing a matching effect that was correlated with schema strength (Experiment 1). This effect was causally driven by schema strength: Weakening group-related schemas through explicit descriptions attenuated the matching effect (Experiment 2). Critically, the effect was domain specific: When social relationships were replaced with nonsocial connections (airport-flight networks), the matching effect disappeared despite comparable structural properties and task demands (Experiment 3). Computational modeling using the Successor Representation framework further revealed that schema-structure misalignment reduced multistep abstraction, reflected in lower successor discount parameters (γ), yielding less integrated global representations of social networks (Experiment 4). Together, these findings show that social network learning is guided by top-down group-related schemas. It advances our understanding of social network cognition by highlighting the importance of schema-driven abstraction and by bridging local learning mechanisms with global representational structure. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

Understanding the intrinsic stage-transition kinetics of lithium-graphite intercalation compounds is central to elucidating the electrochemical performance of graphite anodes in Li-ion batteries, yet quantitatively resolving how individual staging phases transform into one another in real time remains experimentally challenging because neighboring staging phases possess closely related structures and compositions, and their transient coexistence is difficult to deconvolute with sufficient temporal resolution. Here, we establish a stage-resolved kinetic metrology based on time-resolved, in situ 13C magic-angle-spinning solid-state NMR of 13C-enriched graphite, enabling direct, quantitative tracking of the evolution of LixC6 phases during chemically driven delithiation. The large stage-dependent 13C chemical-shift dispersion, combined with the >150-fold signal-to-noise enhancement afforded by isotope enrichment, allows minute-scale acquisition and robust spectral deconvolution of coexisting stage-1 (LiC6), stage-2 (Li0.5C6), and dilute-stage (Li0.33C6) components. Under quasi-equilibrium oxidative delithiation, staging proceeds predominantly through sequential two-phase transitions, LiC6 → Li0.5C6 and Li0.5C6 → Li0.33C6, each well described by Johnson-Mehl-Avrami-Kolmogorov kinetics, consistent with diffusion-limited phase-boundary propagation. This kinetic analysis identifies the dense-stage LiC6 → Li0.5C6 transformation as the intrinsic kinetic bottleneck. When the balance between surface Li removal and intraparticle Li redistribution is perturbed, the staging pathway becomes overlapping and heterogeneous, leading to early emergence of higher-stage phases and extended multiphase coexistence. In these regimes, an effective-order cascade model quantitatively captures the coupled evolution of successive stage transitions. These results reveal how intrinsic stage-transition kinetics and transport constraints jointly govern homogeneous versus heterogeneous delithiation in graphite, and provide a general NMR-based framework for time-resolved quantification of staging transformations in intercalation materials.

ABSTRACT Andersen’s work on self-concepts advanced our understanding by engaging more complex representational forms to better understand how people perceive themselves in relation to significant others in their lives. In the years since, there have been significant advances in our understanding of relational concepts. This paper explores how work on role-governed categories may deepen our understanding of the relational self. Role-governed categories are concepts defined by the roles entities play within relational structures – such as “husband” or “wife” within a marriage – rather than by their intrinsic features. Unlike feature-based categories organized around prototypes, role-governed categories are organized around ideals. We argue that this framework offers three specific advances for relational self-theory: it replaces prototype-based with ideal-based representations of relational partners; the principle of role‑filler independence illuminates how past others become internalized into the self; and embedding social identities within role-governed schemas explains how violated role expectations destabilize self-concept. We discuss implications for self-concept research, social identity, and clinical practice.

ABSTRACT This study conceptualises latent communicative figurations, which are quasi-visible relational structures formed through social media affordances, for example as provided by partnership advertisement systems. We argue that latent communicative figurations are facilitated by platform affordances that enable the curation of visibility, allowing actors to form latent ties for mobilising partners to create and publish social media content that serve specific strategic interests. A systematic approach is proposed to map the positionality of two types of actors – partners (investors) and pages – who contribute to latent communicative figurations, situated between platform designers and the broader networked publics. A sample of branded Facebook posts related to renewable energy is used to construct four bi-partite networks, revealing the structure of latent relations and dominant actors within them. The results show the emergence of a primarily organisational and professional set of latent ties, which mobilises public reactions surrounding renewable energy. The results evidence a dual process of deinstitutionalisation and reinstitutionalisation where actors such as news media transition from their conventional roles to those that are demanded by profit-driven platform logics. The tensions inherent in this transition – where actors must balance the expectations of strategic partners against the public interest – reveal the fundamental antagonisms within networked social environments.

Relational databases are one of the most common data storage mechanisms across many business domains including healthcare, banking, e-commerce, logistics, and human resources. Traditionally, data science teams must join complex relational database structures into a single table, perform manual feature engineering, and train a single machine learning model, per task. This prevents the exploitation of relational signal and increases the required resources needed to scale across business use cases. There have been significant advancements in foundation model research across computer vision, natural language processing, and tabular deep learning. However, there has been limited work, such as Relational Transformer, exploring models that can directly predict on relational databases. Previous work has applied In-Context Learning (ICL) on tasks that have simpler homogeneous data. We leverage advancements in tabular foundation models (TFM), such as TabPFN and ConTextTab, to directly perform ICL on multi-modal relational data. Specifically, we construct a heterogeneous graph via primary and foreign keys in a relational database. We then apply heterogeneous GraphSAGE model as a fixed random feature map to aggregate neighborhood information across the subgraphs. Additionally, we augment entity node representations by combining structural information with the entity embedding. Finally, we supply a TFM with the contextualized entity node representation to perform ICL on arbitrary tasks - without any training. Our method is tested on the public relational deep learning benchmark Relbench, which contains many diverse, real-world predictive regression and classification tasks across healthcare, e-commerce, marketing, and enterprise sales. In a fully ICL regime, our model is shown to be competitive with several fully-trained benchmarks on classification tasks. On average, the method achieves 114% of the fully-trained LightGBM, 95% of the fully-trained relational deep learning model reported in Relbench, and 98% of the zero-shot performance of pretrained RT. This signifies an advancement in unlocking accurate predictions, without training, on the dominant business data structure across diverse domains.

Human attention is often guided by contextual cues that signal where or how to focus in complex environments. Research on object contextual cueing has largely emphasized spatial regularities, but it remains unclear whether socially meaningful information can also provide such guidance. Across three experiments using face stimuli, we examined whether social cues can serve as contextual cues independently of spatial arrangements. In all experiments, participants searched for a target face among distractor faces, with spatial configurations fully randomized on every trial, and contextual information was defined by different types of social mappings. In Experiment 1, the cue was defined by consistent associations between a target identity and a specific set of distractor identities. In Experiment 2, the cue was the overall mood, quantified as the ratio of angry to happy faces. In Experiment 3, the cue was relational information, specifically facing-direction patterns within pairs of profile-view faces. Performance under Consistent Mapping conditions was compared with Variable Mapping conditions, in which these social regularities changed across trials. Bayesian analyses revealed reliable contextual cueing effects across all three experiments, with faster responses under Consistent than Variable Mapping conditions. The effect was strongest in Experiment 2, suggesting that global social cues may be encoded more efficiently than local relational cues. Post-experiment awareness tests indicated minimal explicit knowledge and no reliable association between awareness measures and contextual cueing. Together, these findings suggest that socially meaningful relational structure can provide stable predictive information that supports object-based contextual learning even when spatial regularities are absent.

Antimicrobial resistance (AMR) poses a major threat to global health, and food production environments are increasingly recognized as potential reservoirs and dissemination points for resistant bacteria and antimicrobial resistance genes (ARGs). Slaughterhouse wastewater contains complex microbial communities originating from multiple animal sources and processing activities, yet the effectiveness of current treatment processes in mitigating microbiological and resistome-associated risks remains poorly understood. In this study, we applied high-throughput long-read metagenomic sequencing to characterize microbial community composition and resistome profiles in wastewater samples collected before and after physicochemical treatment from four Italian slaughterhouses. Taxonomic profiling revealed a diverse microbiome dominated by Bacillota and Pseudomonadota, along with DNA assigned to potentially clinically relevant taxa, including members of the ESKAPE group. Resistome analysis identified 96 ARGs conferring resistance to 16 antimicrobial classes. Comparative analyses of pre- and post-treatment samples showed no significant changes in microbial community structure, alpha- and beta-diversity metrics, or ARG profiles. These findings indicate that the applied coagulation-flocculation-based treatment has limited effects on the relative composition of the wastewater microbiome and resistome, as detected by shotgun metagenomics. Our results suggest that slaughterhouse wastewater may act as a persistent environmental reservoir of antimicrobial resistance determinants and highlight the need for enhanced treatment strategies and resistome-oriented surveillance within a One Health framework. Given the limited sample size and the preliminary nature of this investigation, these findings should be interpreted as exploratory and hypothesis-generating, rather than broadly generalizable.IMPORTANCEAntimicrobial resistance is a growing global health concern that extends beyond clinical settings into agricultural and environmental systems. Slaughterhouses represent critical interfaces where microbial communities from livestock, processing environments, and wastewater converge, creating opportunities for the persistence and dissemination of antimicrobial resistance genes. Despite the widespread use of physicochemical treatments to reduce organic load and suspended solids in slaughterhouse wastewater, their impact on microbial communities and resistome remains poorly characterized. By applying long-read metagenomic sequencing, this study provides a comprehensive characterization of the microbiome and resistome in slaughterhouse wastewater before and after treatment. Our findings show that commonly applied coagulation-flocculation treatments do not substantially alter the relative structure of microbial communities or the diversity of resistance genes. These results highlight the potential role of slaughterhouse wastewater as an environmental reservoir for antimicrobial resistance and emphasize the need for improved treatment technologies and systematic surveillance strategies to mitigate the environmental dissemination of resistance determinants in line with the One Health approach.

Toxoplasma gondii is a globally important intracellular parasite, and treatment regimens are limited by the failure of drugs to target latent tissue cysts. Developing new candidates for treatment also needs to address the potential for resistance to arise. Herein, we developed a minimum inoculum for resistance assay as a semiquantitative metric for evaluating inhibitors of T. gondii. The resistance assay, adapted from malaria, measures the frequency of pre-existing resistance alleles by exposing different-sized parasite populations to drug pressure. We profiled a series of bicyclic pyrrolidone analogues that inhibit phenylalanine tRNA synthetase. We demonstrate that these inhibitors require higher inocula to lead to parasite resistance (up to >108 parasites) in comparison with an inhibitor of DNA synthesis and that resistance values vary across inhibitors with closely related chemical structures. Clonal analysis of resistant parasites emerging from resistance assays revealed both new and previously identified resistance-conferring mutations in T. gondii phenylalanine tRNA synthetase, and structural modeling revealed their potential impact on the enzyme active site. The minimum inoculum for resistance assay provides a functional benchmark to compare new and existing inhibitors, allowing for rational prioritization of lead compounds with a high genetic barrier to resistance.

Altered immune states at the time of cancer diagnosis remain insufficiently characterized. Although circulating immune biomarkers offer a promising, non-invasive way of analysing systemic tumour–host interactions, their potential remains poorly defined. Here, we present an integrated multi-omics analysis of peripheral blood mononuclear cells from treatment-naïve cancer patients, minimizing confounding by therapy-induced immune changes, combining immune phenotyping (flow cytometry, FC), multiplex cytokine profiling, and single-cell RNA sequencing (scRNA-seq). Compared with healthy donors, patients exhibited widespread immune dysregulation, including expansion of FOXP3+ regulatory T cells, depletion of CD16+CD11b+ monocytes and CD56^dim^ Natural killer (NK) cells, and elevated plasma IL-6 and IL-4 levels. scRNA-seq identified cancer-associated immune signatures, notably consistent upregulation of THBS1 and CH25H, indicative of systemic imprinting by tumour-derived cues. We further developed machine learning-guided models integrating single-cell multi-omics data (sc-FC and scRNA-seq) to characterize cancer-associated immune patterning and cancer type–related signal structure, while providing biologically interpretable feature attribution across modalities. The models achieved robust classification performance within the cohort and revealed modality-spanning features linked to immune state alterations.Together, these findings establish a framework for immune-based, multi-omics profiling of peripheral blood and provide a resource for discovering circulating cancer-associated immune signatures. This supports future development of immune-based diagnostics and disease monitoring approaches.

Integrating diverse biomedical modalities is essential for robust healthcare insights, and graph-based models are increasingly used to capture complex relational structures. Yet, their clinical translation hinges on interpretability. This review surveys interpretable graph-based models applied to multimodal biomedical data, highlighting dominant trends in disease classification, static graph construction, and post-hoc explainability. We categorize explainable artificial intelligence (XAI) techniques, benchmark SHAP, saliency, sensitivity, and graph masking on Alzheimer's disease data, and reveal complementary strengths. A development flowchart and future directions, such as dynamic graphs, knowledge integration, and LLM-based explainability, position this work as a key reference for trustworthy biomedical AI.

Musculoskeletal disorders (MSDs) represent a diverse group of conditions affecting bones, cartilage, intervertebral discs, tendons, muscles, and related structures, and continue to be a leading cause of global disability. Despite advancements in surgery and pharmacotherapy, issues such as incomplete regeneration, limited longevity, and procedure-related complications remain unresolved. As a result, hydrogel technologies have garnered increasing attention due to their hydrated three-dimensional networks, customizable mechanical properties, and extracellular matrix (ECM)-mimicking features. Hydrogels have been developed in various forms, including injectable systems, scaffolds, and patches, to facilitate tissue regeneration, deliver bioactive molecules, modulate inflammation, and provide mechanical or interfacial support. This review comprehensively examines recent advancements in hydrogel technologies for MSD treatment, focusing on bone and cartilage repair, osteoarthritis management, intervertebral disc regeneration, tendon and muscle healing, and spinal dural repair. It covers laboratory studies, preclinical animal trials, and clinically approved or translational products. By critically evaluating representative materials, design strategies, and disease-specific requirements, this review highlights the therapeutic potential and translational barriers of current hydrogel systems. Challenges such as mechanical durability, spatiotemporal control of bioactivity, scalable manufacturing, and clinically relevant evaluation are addressed, providing insights into the development of next-generation hydrogels for MSDs.

ABSTRACT This paper examines the social and cognitive antecedents of sexual harassment-related behavioural climate. Drawing on the theory of planned behaviour and social identity theory, this paper establishes relationships between social identity, homophily, social ties, customer-employee interaction, and attitudes toward sexual harassment. Data were collected from 589 employees working at three-star hotels in Egypt. Results showed that social and cognitive dynamics in hospitality workplaces are key determinants of attitude towards sexual harassment, forecasting acceptance of sexual harassment-related behavioural climate. This paper adds to the hospitality literature by theorising sexual harassment as a socially constructed behavioural climate and highlighting the importance of relational structures in affecting employee perceptions and workplace behaviours.

The article carries out a comprehensive criminological study of persons who commit criminal offenses in the sphere of activity of non-banking financial institutions. It is substantiated that the social danger of such offenses is largely determined by the set of socio-demographic, status-role, moral-psychological and criminal-legal characteristics of the offender’s personality, which are formed under the influence of the social environment and are implemented in illegal behavior. It is determined that the personality of the criminal is a complex interdisciplinary phenomenon that combines social, psychological and legal features and is considered as a carrier of anti-social orientation. It has been established that persons who commit criminal offenses in the field of non-banking financial institutions are characterized by a high level of education, the presence of special knowledge in the field of finance and lending, professional competence, as well as developed communication skills, the ability to persuade and manipulate. It has been proven that their behavior is mainly rational, self- motivated in nature and is implemented through the use of official position, trust relationships and gaps in legal regulation. It is substantiated that criminal offenses in this area are mostly organized in nature, committed in groups with a clear division of roles and functions, which makes their detection and investigation difficult. The author’s approach to the classification of offenders is proposed, which involves their division into internal, external, associated and intermediary entities depending on the degree of involvement in financial and credit activities. It is emphasized that the key role in the mechanism of criminal activity is played by persons who occupy managerial and supervisory positions in non-bank financial institutions or related structures, since it is they who have access to financial resources, management tools and the ability to influence decision-making. The conclusion is made about the need to improve criminological prevention measures taking into account the typological characteristics of the offender, as well as strengthening internal control, compliance and regulatory supervision in the financial sector.

Traditional fuzzy and rough set models usually have a hard time in providing sufficient ambiguity, hesitancy, and partiality to information that is common in actual hospitality decision-making. In order to overcome these problems, this paper suggests a new decision-support model that is built upon pessimistic multi-granulation rough sets combined with cubic intuitionistic fuzzy soft relation. The model proposed is a unification of interval-valued membership, intuitionistic non-membership and soft binary relations in a structure of pessimistic rough approximations. The framework characterizes multi-dimensional uncertainty of complex hospitality appraisals by modelling membership, non-membership, hesitation and boundary regions together in a variety of granulations. Soft binary relations defined in terms of foresets and aftersets are used to form lower and upper pessimistic multi-granulation approximations of internal cubic intuitionistic fuzzy sets. Approximation operators are formally defined in two pairs and the algebraic property of the operators is explored. In addition, measures of similarity among internal cubic intuitionistic fuzzy sets on soft relational structures are created to facilitate the robust alternative comparison. On the grounds of these theoretical bases, a systematic decision-making scheme that is composed of two structured algorithms is set up. The relevance of the suggested framework can be proved by a practical hotel selection case study with the usage of several factors and professional evaluations. Comparative and sensitivity analysis demonstrates that the model proposed gives consistent rankings and minimizes the effects of boundary ambiguity in different levels of uncertainty, which can be used as a trusted and risk-conscious decision support device in hospitality management.

ABSTRACT This article examines KissCross, Thailand’s first virtual Boys’ Love idol duo, who transitioned from webtoon characters to nationally broadcast performers within six months (June to December 2025). Drawing on platform data, social media documentation, and discographic analysis, the article introduces ‘curated proximity’ to describe intimacy engineered to feel personal whilst systematically filtering biographical exposure. Situating KissCross within the broader landscape of virtual idol production, including Japan’s Hatsune Miku and South Korea’s PLAVE, the analysis distinguishes between performance-origin virtual idols, whose authenticity emerges through what surfaces during performance, and narrative-origin virtual idols, whose intimacy rests on completed fictional canon. Rather than overturning parasocial relationship theory, curated proximity extends it by identifying a modality in which distance operates not as deficit but as preferred relational structure.

Recent work has shown that people are sensitive to coarse differences in network topology, including network features like "holes," "crosses," and "T-junctions." Even children as young as 4years old will readily distinguish between items that differ slightly in their network topology. But how robust is this sensitivity? Here, we evaluate whether people are not only sensitive to differences in the presence or absence of certain topological features, but also to their exact spatial arrangement. In a first experiment, we show that people distinguish figures which possess all the same topological features as other figures in a set if the features differ in spatial arrangement. In a second experiment, we show that people also match figures based on exact spatial arrangement. Finally, we show that memory encodes the correct relational structure of the figures: People are more likely to falsely indicate having seen an item if it shared the precise arrangement of topological features of other items they had seen (compared to a closely matched item which had the same features arranged in a different way). Combined, these results bolster the theory that people intuitively appreciate the precise spatial arrangement of topological features.

ABSTRACT This article examines how planning interventions implemented in Galata after the 1950s transformed street–urban block relationships, using the spatially directive effects of the wall system as an analytical framework. While the study is conceptually informed by the historico-geographical approach, it does not present a conventional historical geography application nor a normative urban plan analysis. Instead, it adopts a cartography-based place-reading framework, conducting a comparative analysis of the 1944 Topographical and Archaeological Plan of Galata and the current 2024 digital plan. The primary analytical scale of the study is the street and urban block. The findings reveal that in the 1944 urban fabric, street–block relationships exhibited strong morphological continuity shaped by the persistent spatial influence of topography and the wall system. In contrast, transport-oriented planning decisions, road widenings and coastal rearrangements disrupted this relational structure, producing diverse morphogenetic patterns such as fragmentation, selective transformation and spatial resistance. Through the case of Galata, the study argues that urban transformation in historic port cities should be interpreted not through a singular narrative, but through spatially differentiated morphogenetic processes.

ABSTRACTBackgroundMachine learning models for stroke mortality prediction typically treat each time horizon independently and use flat tabular features that ignore the relational structure of electronic health records (EHRs). In this pilot study, we leveraged graph-based machine learning models to predict post stroke all-cause-mortality across three different time horizons.MethodsWe developed Stroke Temporal Heterogeneous Graph (StrokeTHG), a heterogeneous graph neural network model for simultaneous multi-horizon stroke mortality prediction (30-day, 90-day, 1-year) using EHR data from Penn State Health System. The model encodes various relations among EHR entities (e.g., patient, diagnosis, comorbidity) and temporal encoding of admission time to better predict stroke mortality. We compared our proposed approach against various baseline methods, including Logistic Regression, Random Forest, and XGBoost. We also performed ablation and subgroup analyses, evaluated the quality of learned graph embeddings, and assessed the importance of different edge types in the graph.ResultsWe included 4,144 stroke patients (mean age 69.2 years; 54.3% men), of whom 3,332 (80.4%) survived their stroke after one year. 30-day, 90-day, and 1-year mortality rates were 9.7%, 13.7%, and 19.6%, respectively. Our proposed approach, StrokeTHG, achieved AUROC of 0.872, 0.878, and 0.837 across horizons, outperforming all tabular baselines. At ≥75% specificity, the model identified 5–10 percentage points more mortality cases than the best baseline at each horizon. Subgroup analysis demonstrated consistent performance across sex subgroups and the largest discriminative gains in the Age 65–80 stratum. Edge-type ablation identified phenotype–patient and admission–patient edges in the constructed EHR graph as the most influential relational edges for mortality prediction. StrokeTHG embeddings outperformed all graph and matrix factorization baselines under an identical downstream classifier, confirming that performance gains stem from representation quality rather than classifier capacity.ConclusionsStrokeTHG demonstrates that heterogeneous graph representations of EHR data provide a consistent improvement over flat tabular models for multi-horizon stroke mortality prediction, with particular advantage at clinically actionable sensitivity thresholds and novel multi-horizon monotonic prediction capability. This methodological framework may be adaptable to other EHR-based clinical research studies seeking to leverage heterogeneous relational structures for predictive modeling.

Musical modes (i.e., different patterns of pitch organisation within a scale) are widely recognised for their emotional character, yet little is known about the broader multisensory associations that listening to them might trigger. In the present exploratory study, we examine whether the full set of Western musical modes elicits systematic patterns of visual and olfactory mental imagery and whether these patterns correspond with how listeners spontaneously group the modes. In all, 249 participants generated open-ended visual and olfactory responses while listening to short modal excerpts and subsequently completed a free-sorting task. The results revealed a number of structured and recurrent imagery themes across participants (e.g., Nature-, Daytime-, and Happiness-related imagery for major modes, versus Dark-, Stress- and Sadness-related imagery for minor modes, alongside Floral-and-Fresh- versus Damp-, Dusty- and Smoke-related olfactory associations). Major and minor modes occupied distinct regions of similarity space across both visual and olfactory modalities; however, meaningful differentiation was also evident at the level of individual modes. This consistent relational structure across visual and olfactory imagery was likewise reflected in the grouping patterns observed in the free-sorting task. Together, these findings indicate that musical modes are associated with multiple sensory representations that extend beyond simple feature-level correspondences. These results therefore provide an exploratory mapping of visual and olfactory mental imagery and establish a foundation for future confirmatory research.