Development and validation of an eye-tracking-based cognitive impairment screening system for older adults in China: a cross-sectional study.

Large Language Models (LLMs) can serve as tools for understanding how probabilistic constraints interact during language acquisition. To motivate such use cases of LLMs, we discuss several examples from allied fields, including neurobiology and animal behavior, of how soft constraints shape learning and development in cognitive systems. We end by outlining four challenges that LLM cognitive modeling should address in the coming decade.

Mammalian sensory systems are traditionally viewed as hierarchical pathways in which subcortical nuclei relay signals from peripheral receptors to the cortex, where sensory information is contextualized for perception and behaviour. However, the auditory pathway contains an unusually large number of anatomically complex, recurrently connected subcortical nuclei that transmit heavily pre-processed information to the cortex. Emerging evidence shows that the auditory system functions as an integrated cortical-subcortical networkthat is heavily influenced by non-auditory inputs and extensive bidirectional connectivity at nearly all levels. Subcortical structures are not passive relays but active participants in computations traditionally attributed to cortex, including adaptive coding of sound statistics, multisensory integration, encoding of behavioural relevance and action, and learning. Although certain transformations occur hierarchically - such as brainstem spatial processing - sensitivity to most sound features forms a continuum across the auditory pathway, with modulation by other sensory, motor and cognitive systems at every stage. The degree of subcortical pre-processing may explain why emergent cortical properties are harder to identify in audition than in vision, in which cortical and subcortical receptive field properties are more distinct. This distributed circuit organization opens new avenues for understanding how the brain constructs perception and guides behaviour from the fusion of sensory evidence and contextual knowledge.

The rhythms of trance: Cultural phenomenology and neural mechanisms of music-induced non-ordinary states of consciousness.

From peripersonal space to cognitive maps: An evolutionary perspective.

Objective: this article examines the feasibility of implementing an active safety and driver assistance system (ASDS) on a city tram. Methods: analysis of manufacturer data and the operator’s experience using the Cognitive TramPilot system on tram rolling stock in St. Petersburg. Results: implementation of this system has reduced the number of tram derailments, passenger falls, pedestrian coll isions, and vehicle-vehicle accidents. Practical significance: installation of this system on all Gorelektrotrans rolling stock can be recommended to improve traffic safety and transportation efficiency, and, ultimately, to facilitate the transition to driverless trams.

Poststroke aphasia (PSA) significantly impairs language and cognitive functioning, yet few interventions comprehensively address both domains. This study evaluated the efficacy of an intervention combining acupuncture and computer-assisted cognitive rehabilitation using the RehaCom cognitive training system (hereinafter, ACR) on language and nonverbal cognitive function in individuals with PSA. A single-center randomized controlled trial was conducted at a tertiary rehabilitation hospital in Fujian Province, China. Eighty patients with PSA were randomly assigned (1:1) to either the combined-intervention group (ACR plus standard speech and language therapy [SLT]) or the SLT group (SLT only). The intervention consisted of five 30-min sessions per week. Primary and secondary outcomes were assessed using the Western Aphasia Battery-Revised (WAB-R) and the Non-language-based Cognitive Assessment (NLCA) at baseline, Week 4, and Week 12 (post-intervention). Generalized estimating equations were used to evaluate group differences over time. Of the 80 enrolled participants, 71 (88.75%) completed the 12-week intervention. Compared with the SLT group, participants in the combined-intervention group demonstrated significantly greater improvements in the WAB-R Aphasia Quotient at both 4 weeks (B = 5.961, 95% CI [3.765, 8.158], p < .001, Cohen's d = 0.135) and 12 weeks (B = 11.806, 95% CI [8.050, 15.563], p < .001, Cohen's d = 0.460). Significant gains were also observed in WAB-R subdomains, including Spontaneous Speech, Auditory Comprehension, Repetition, and Naming, with small-to-moderate effect sizes. Moreover, the combined-intervention group exhibited substantial improvements in nonverbal cognitive function, as measured by the NLCA, with moderate-to-large effect sizes at both post-intervention points (Cohen's d = 0.434 at 4 weeks; Cohen's d = 0.847 at 12 weeks). The ACR intervention yielded significant and clinically meaningful improvements in both language and cognitive functions in patients with PSA. This multimodal approach represents a promising therapeutic strategy for comprehensive aphasia rehabilitation.

The rapid expansion of urban air mobility operations demands adaptive airspace management approaches that transcend traditional static sectorization. This paper proposes an integrated framework for dynamic low-altitude airspace partitioning and management strategy optimization by fusing graph neural networks with spatial cognitive science. Urban low-altitude airspace is modeled as an attributed weighted directed graph encoding spatial adjacency, traffic flow coupling, and environmental constraints. A spatial cognitive constraint system is developed, quantifying boundary discriminability, shape complexity, and hierarchical cognitive load as differentiable optimization terms. A cognition-enhanced graph attention network architecture with spatiotemporal feature aggregation is designed to generate end-to-end partition assignments, with an explicit cognitive attention modulation mechanism that steers message-passing toward perceptually coherent neighborhoods. A reinforcement learning module, formalized as a Markov decision process with clearly defined state, action, and reward spaces, fine-tunes the partition and management strategy outputs through proximal policy optimization. The joint training scheme co-optimizes sector boundaries with capacity allocation, priority sequencing, and conflict alert policies through multi-task learning with curriculum scheduling. Experimental results on both synthetic simulation data and real-world ADS-B trajectory data from the OpenSky Network demonstrate that the proposed method achieves an airspace utilization rate of 81.2 ± 1.4% and reduces the flight conflict rate to 3.1 ± 0.5 per 100 flight-hours, representing a 46.6% improvement over vanilla graph attention baselines and outperforming advanced spatiotemporal GNN baselines including STGCN and DCRNN. Statistical significance is confirmed via Wilcoxon signed-rank tests (p < 0.01). A small-scale human-in-the-loop study with eight certified air traffic management researchers points to a roughly 37% reduction in operator decision response time, a signal we read as preliminary rather than conclusive given the modest sample. Scalability tests indicate that model inference alone stays feasible up to 10,000-node airspace graphs within the reconfiguration window, though we are careful to note that the complete operational pipeline has not yet been integrated end-to-end.

Untangling relationships between cognitive development and child and adolescent mental health: Findings from the ABCD Study.

The Flatland thought experiment, drawn from Abbott's 1884 novella and developed by Carl Sagan, has been increasingly applied to neurodiversity discourse and artificial intelligence ethics as a metaphor for constrained perception. The standard reading positions neurotypical cognition as the three-dimensional Sphere — the more complete, higher-dimensional observer — and neurodivergent cognition as the two-dimensional Square, generating a cross-section interpreted as disorder rather than difference. This article argues that this reading encodes the deficit model it was designed to challenge and introduces Epistemic Parallax as a novel theoretical construct to correct it. Epistemic Parallax is defined as the systematic displacement in meaning, classification, and judgment that arises when a cognitive system, institutional framework, or artificial intelligence observes neurodivergent experience from a non-parallel normative frame — producing distortions that are a geometric property of the observational relationship rather than a feature of the observed. The construct is distinguished from the Double Empathy Problem, institutional ableism, and algorithmic bias by its specification of mechanism over outcome and its applicability to non-adaptive systems that cannot self-correct through reciprocal interaction. Grounded in the Double Empathy Problem's empirical record, the full dimensional progression from point to tesseract, Intense World Theory, monotropism, and multidimensional sensory processing research, Epistemic Parallax is applied to AI in digital mental health to identify the deployment of neurotypically-trained systems as clinical arbiters of neurodivergent experience as a form of structural hermeneutical injustice in Fricker's precise sense. The Dimensional Parity Standard is proposed as the operational correction, comprising six criteria — bidirectional validation, cross-plane transparency, co-authorship of ground truth, relational deployment, dimensional humility, and a sixth principle extending the Feynman honesty framework developed in the companion article. Implications for regulatory policy, system design, and a three-priority research agenda are identified.

Brain health disorders (BHDs) remain a concern for people with HIV (PWH) despite antiretroviral therapy access and viral suppression. The contribution of HIV to brain health is often obscured by comorbidities in high-income settings which are less prevalent in sub-Saharan Africa. Neurofilament light chain (NfL), a biomarker of axonal injury, may offer insight into underlying mechanisms. 338 virally-suppressed PWH and 250 people without HIV (PWoH) completed a Research Domain Criteria-informed battery assessing cognitive, sensorimotor, and social processing systems. Demographically-adjusted norms were derived from PWoH. Serostatus differences in impairment (≥ 1SD below the mean) were examined using multivariable logistic regression. Additional models examined associations between NfL (plasma, cerebrospinal fluid [CSF]) and task performance. PWH were similar to PWoH in age (43.9 vs. 43.5yrs), sex (female, 54 vs. 46%), and education (6.1 vs. 5.8yrs). PWH had higher odds of impairment in the cognitive control and attention (Color Trails, Symbol Digit) and sensorimotor (Grooved Pegboard) domains. Plasma NfL was associated with sensorimotor impairment in both groups. Similar trends held in CSF NfL but did not reach statistical significance, likely due to sample size (n = 85). Cognitive and sensorimotor difficulties are common in PWH in Rakai, independent of typical Western confounders. The profile of impairment differs from reports in high-income settings where declarative memory deficits are often observed. NfL was associated with sensorimotor impairment, suggesting that NfL may capture ongoing axonal injury and motor system vulnerability in PWH and PWoH. These findings suggest NfL's potential as a biomarker of sensorimotor impairment in sub-Saharan Africa.

Psychopathy is a construct characterised by multiple regulatory pathways, mental representations, emotional expressions, and behaviours. Although assessment tools retain significant clinical and forensic value, they focus on enduring traits and overt behaviours. These measures tend to overlook the dynamic and context-dependent processes that also influence psychopathic functioning. This article proposes that the Cognitive-Affective Processing System (CAPS) offers a framework for integrating both the stable and dynamic elements of psychopathy. CAPS conceptualises personality as a system of cognitive-affective units that interact with situational cues to produce characteristic ‘if–then’ behavioural signatures. The low-arousal hypothesis, the response modulation theory, and the hot–cold framework within CAPS clarify the distinctions among psychopathic phenotypes. This enhanced meta-model also has the potential to explain dormant periods in serial offenders and sex differences in psychopathy. Rather than replacing existing measures, the CAPS meta-model improves on the interpretive value of other measures and facilitates forensic decision-making.

Explainable Artificial Intelligence (XAI) is gaining popularity in early diagnosis and monitoring of dementia. Herein, we recommend the incorporation of the National Institute of Mental Health's Research Domain Criteria (NIMH-RDoC) framework with XAI-informed diagnostic protocols to help establish diagnosis at early stages of Alzheimer's disease (AD). RDoC has a dimensional structure that extends across units of analysis from genes and molecules to circuits, physiology, behavior, and introspection. By restructuring diverse features as inputs including apolipoprotein E (APOE) genotype, amyloid and tau biomarkers, computational neuroimaging-informed cortical atrophy, Positron Emission Tomography (PET) hypometabolism, quantitative electroencephalography (qEEG) rhythms, cognitive tests, and digital behavioral markers), onto RDoC units provides more insightful and inclusive models. In this context, data-driven approaches such as XAI can achieve not only increased interpretability but also enhance their mechanistic validity. Such an innovative approach places data-driven model outputs within neurobiologically based domains such as Cognitive Systems, Negative Valence, and Arousal/Regulatory Systems. Our synthesis suggests that a 'converging RDoC and XAI' approach may help bolster the coherence of AD biomarkers, promote model exploration in clinical decision-making. This approach is also expected to provide a strategic roadmap for translational neuroscience and personalized medicine. Another major aim of this study is to critically analyze current XAI approaches used in dementia research, particularly the diagnostic and prognostic aspects. By explicitly grounding explanations in RDoC cognitive domains and paradigms, the framework also aims to make model outputs meaningful in terms of specific mental functions (e.g., episodic memory, cognitive control), thereby supporting neuropsychologically-informed diagnosis, categorization, and communication with patients and caregivers.

Recent advancements in digital transformation and agile methodologies have significantly reshaped how modern organizations manage complex workflows. Effective project management and seamless team collaboration are now recognized as essential pillars for maintaining productivity in high-stakes environments, including software development, healthcare coordination, and remote corporate operations. Traditional task management systems often struggle with fragmented communication and siloed data, leading to reduced system efficiency, increased operational complexity, and a lack of real-time synchronization between team members. This paper proposes an innovative, integrated Task Management and Collaboration Tool designed to streamline organizational workflows through a unified digital framework. The system architecture is designed to address the critical limitations of existing fragmented tools by providing a centralized hub for all project-related activities. Utilizing advanced software architectures, the platform enables real-time tracking, intelligent task prioritization, and multi-user interaction within a single interface. The proposed methodology incorporates automated status updates and centralized data repositories for resource sharing, ensuring that all stakeholders have access to the most current project metrics. Furthermore, the system is engineered for cross-platform accessibility, ensuring robust performance and scalability across diverse and demanding professional settings. By integrating these features, the tool minimizes the need for multiple independent applications, thereby reducing cognitive load and system overhead.

Repetitive soccer heading has been implicated as a potential source of cumulative subconcussive brain injury, yet the magnitude and consistency of its cognitive effects remain incompletely defined. We conducted a systematic review and meta-analysis to quantify global and domain-specific cognitive outcomes associated with repetitive heading exposure. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guidelines, PubMed/MEDLINE, Scopus, and Web of Science were searched through December 2025. Of 2,846 identified records, 46 studies met the inclusion criteria, and 28 provided standardized cognitive data suitable for quantitative synthesis. Effect sizes (Hedges g, Fisher z) were transformed into log odds ratios and pooled using DerSimonian-Laird random-effects models. Heterogeneity was assessed using Q, I2, and H2 statistics, and publication bias was evaluated with funnel plots, Egger regression, and trim-and-fill procedures, alongside leave-one-out influence analyses. Across 28 independent author-level datasets, repetitive heading was associated with significantly increased odds of global cognitive underperformance (odds ratio 1.67; 95% confidence interval 1.61-1.72), with moderate heterogeneity (I2 ≈ 34%). Trim-and-fill adjustment yielded a modestly attenuated but still significant estimate (odds ratio 1.49). Domain-level analyses demonstrated consistent impairments across visuospatial ability (odds ratio 1.49), verbal memory (odds ratio 1.62), attention (odds ratio 1.71), processing speed (odds ratio 1.64), executive function (odds ratio 1.86), and composite cognition (odds ratio 1.58). Confidence intervals were narrow, once, and the effect directionality was uniform across domains. These findings indicate that repetitive soccer heading is associated with robust, reproducible cognitive deficits across multiple cognitive systems, supporting cumulative subconcussive exposure as an independent risk factor for measurable cognitive decline.

This paper examines the clinical gap between identifying neuroplastic pain and achieving treatment success with Pain Reprocessing Therapy (PRT), proposing a structured readiness framework to guide clinicians in determining when patients are appropriately positioned to engage with this intervention for chronic primary pain. Despite a randomized controlled trial demonstrating that approximately 66% of PRT participants achieved pain-free or nearly pain-free status at one year, no implementation guidelines exist for establishing patient readiness prior to initiating treatment. This paper proposes ten conditions across four domains: patient cognitive readiness, patient behavioral readiness, provider alignment, and system-level factors. Evidence from pain neuroscience education, the fear-avoidance model, stages of change literature, and therapeutic alliance research supports the theoretical grounding of each condition. Emotional Awareness and Expression Therapy is proposed as an adjunctive intervention for patients whose pain is sustained by unresolved trauma or emotional inhibition beyond the reach of cognitive reattribution alone. Neuroplastic pain identification is necessary but may be insufficient for PRT success. Unmet readiness conditions spanning cognition, behavior, provider practice, and healthcare systems may explain a meaningful proportion of limited treatment responses. These proposals require empirical validation before clinical adoption and should be treated as hypotheses to guide implementation research rather than established practice guidelines.

This research note discusses a collaborative scholarship model integrating funded research with Community Engaged Learning (CEL) between undergraduate courses at Michigan State University and Kansas State University. Cross-institutional student groups comparatively assessed national forest management issues in Southern California with those across the Midwest. Guided by CEL principles, students interacted virtually with Forest Service managers, engaged in experiential learning, and practiced leadership. This model bridged research and education and fostered professional skills and community building across institutions. Outcomes suggest that this model is adaptable for other courses. Key student outcomes include cognitive (systems thinking, forest management), affective (empathy, civic identities), and professional (communication, leadership, resilience) aspects. Key teaching outcomes include the collaborative model’s replicability and adaptability, careful strategic planning across communities and throughout the experience, successful student preparation and support on process and product, utility of student reflections for student and instructor growth, and cross-course enrichment deepening meaningful pedagogy.

ABSTRACTEnvironmental input is inherently uncertain at both global and local levels. Depending on the level of uncertainty, cognitive systems dynamically adapt by shifting attention between exploiting learned regularities and exploring less predictable alternatives. While these dynamics are well documented in adults, how local and global uncertainty jointly influences children's exploration‐exploitation strategies during statistical learning remains unclear. Combining eye‐tracking with a probabilistic cueing task, this study quantified local and global uncertainty on a trial‐by‐trial basis to examine their effects on children's attentional preferences for cues differing in predictive value. Overall, 158 children aged 4 to 13 years passively viewed cue‐target sequences in which target shapes were predicted by cues with high‐ and low‐transitional probabilities. Results showed that under high global uncertainty, high local uncertainty biased attention toward high‐predictive cues, indicating uncertainty‐exploitation, whereas low local uncertainty biased attention toward low‐predictive cues, suggesting certainty‐exploration. Conversely, under low global uncertainty, attentional strategies reversed, shifting toward certainty‐exploitation and uncertainty‐exploration. These patterns were amplified in older children but attenuated in younger ones. These findings support a developmental dynamic model of exploration‐exploitation in which children's attention strategies are jointly regulated by local and global uncertainty and undergo systematic change across childhood.SummaryHow local and global uncertainty jointly shapes the development of children's exploration‐exploitation behavior during statistical learning across different ages remains poorly understood.Using eye‐tracking in a probabilistic learning task, we showed that 4‐13‐year‐olds adapt their information sampling and choices in response to overarching and trial‐by‐trial uncertainty.Under high global uncertainty, children relied on local uncertainty for exploitation and exploration; under low global uncertainty, they favored certainty‐driven exploitation and uncertainty‐driven exploration.These uncertainty‐regulated strategies were amplified in older children but attenuated in younger children.

Conducting dynamic exploration in complex and unpredictable environments, particularly in space exploration, reveals the great potential of systems based on tensegrity structures. However, the implementation of such systems faces a series of intelligent challenges, including the reliability of wireless monitoring, the efficiency of human-computer interaction, and the optimization of intelligent analysis and recommendation capabilities. In this study, we introduce a 6-bar tensegrity system equipped with 24 flexible sensors, leveraging fine-tuned multimodal large language model to enable autonomous state cognition system including self-shape recognition, alarm system, as well as fault diagnosis. Supported by long and short-term memory models, the tensegrity is able to reconstruct its own shape via conductive flexible tendons without relying on external sensors. By combining the flask server with the fine-tuned large language model, the tensegrity automatically transmits data to the iPhone for wireless monitoring. Finally, we developed the fine-tuned LLM and employed it to facilitate fault diagnosis and human interaction, enabling users to effectively obtain the requisite information through natural language processing techniques. This autonomous state cognition system relying on tensioning bars shows great potential for future exploration and becomes a powerful tool for multifunctional applications in the real world.

Existing research focuses on data-driven algorithm optimization but overlooks the embodied nature of supply chains as physical and digital integrated systems, leading to a disconnect between AI and physical collaboration. This study introduces embodied intelligence into supply chain management, transcending the traditional paradigm to propose an adaptive collaboration framework through embodied perception, contextual reasoning, and physical execution. It deconstructs the core of supply chain embodied intelligence, revealing issues such as fragmented perception and delayed feedback. Based on embodied cognition and complex adaptive systems theory, a four-layer architecture with embodied perception, contextual reasoning, physical execution, and closed-loop feedback is constructed, clarifying its mechanisms. Future directions in theory, technology, and practice are outlined. This work deepens the integration of embodied intelligence with supply chains, bridges the digital and physical divide, and advances supply chain management toward an embodied adaptive paradigm for next-generation intelligent systems.