Search Results (1,548)

Air-to-air heat pumps are a key technology for improving energy efficiency and reducing carbon emissions in residential buildings, yet their optimal control remains challenging under real-world conditions. This study evaluates the performance of a hybrid physical–LSTM model for controlling an air-to-air heat pump in a residential building in Zadar, Croatia. The hybrid framework integrates a first-order energy balance model of the building envelope with LSTM-based temperature correction using adaptive weighting. The physical model was calibrated and validated against 52,128 real IoT measurements collected during the 2024/2025 heating season, achieving high accuracy (RMSE ≈ 0.076 °C). Rolling one-day and continuous multi-day closed-loop simulations (up to 15 days) show that the hybrid model yields slightly lower RMSE in long-term runs compared to the pure physical model. However, this apparent statistical improvement is accompanied by systematic underestimation of indoor temperature and significantly higher simulated energy consumption. The results indicate that the observed effect originates from an implicit virtual heat flux introduced by the LSTM correction, which affects thermodynamic consistency in closed-loop operation. The findings highlight that short-term error metrics such as RMSE alone are insufficient for evaluating hybrid models intended for model predictive control (MPC). The main contribution of this study is the explicit demonstration and quantification of an implicit virtual heat flux generated by the LSTM correction in closed-loop multi-day operation, which leads to misleading statistical improvements while causing significant thermodynamic inconsistency and energy overconsumption. In 15-day continuous simulations the hybrid model (ω = 0.05–0.10) caused an indoor temperature underestimation of 1.25–1.31 °C and increased simulated electricity consumption by more than 300% (316 kWh vs. 72 kWh) compared to the physical model. These results have direct implications for the development of reliable digital twins and model predictive control strategies in residential HVAC systems. Full article

The incorporation of digital technologies has transformed Social Work, generating new demands in terms of professional competencies. It is worth questioning whether, in contexts as sensitive as social services, the mere acquisition of instrumental mastery of these tools is sufficient to ensure safe practice. Considering that the level of proficiency with these tools is influenced by age, the aim of this study is to conduct a comparative analysis of the digital competence levels of social work professionals and students in order to contrast the skills and shortcomings of both groups. To achieve this, a quantitative methodological design was employed using questionnaires based on the European DigComp 2.2 framework. The sample included 451 professionals from Spain and 171 students from the University of Huelva, whose data were processed using statistical software (SPSS 27). The results show that students display a higher overall level of digital competence, particularly in the creation of digital content and the use of artificial intelligence tools. Professionals, on the other hand, demonstrate stronger competencies in digital security and data literacy. The findings reveal a clear complementarity between the two groups, highlighting an opportunity for mutual feedback and learning that can help combine technological efficiency with ethical responsibility and respect for human rights. Full article

Background: Percussive massage therapy (PMT) with handheld massage guns is widely used to support recovery and flexibility, but the short-term behavior of skeletal muscle contractile properties and the relative contribution of application duration versus frequency remain unclear. This study investigated the 10 min post-intervention time course of tensiomyography (TMG)-derived contractile properties of non-fatigued vastus medialis (VM) after clinically realistic PMT protocols and examined whether longer duration is associated with persistent deviations from baseline than frequency. Methods: In a two-session, within-subject repeated-measure design, 32 participants completed four PMT conditions to the VM (35 Hz–3 min, 35 Hz–6 min, 45 Hz–3 min, and 45 Hz–6 min). TMG parameters (Td, Tc, Ts, Tr, and Dm) were recorded at baseline and repeatedly over 10 min post-intervention. Linear mixed-effect models with frequency and duration as fixed factors and time as continuous and categorical were used to characterize temporal patterns, with emphasis on effect sizes and consistency across parameters. The fixed protocol order (35 Hz in session one, 45 Hz in session two, 3 vs. 6 min assigned to contralateral legs) means that frequency was confounded with session and duration with leg side. Results: Compared with the 3 min protocols, the 6 min protocols were associated with slightly higher Td and Ts, a modest increase in Tr and a slightly greater Dm (e.g., Dm + 0.55 mm), whereas Tc showed no clear duration effect. Across conditions, Td increased immediately after PMT, Tc remained elevated for most of the first 8 min, Ts increased from mid to late post-intervention, Tr changed inconsistently, and Dm was reduced relative to baseline for most of the 10 min period. Differences between 35 and 45 Hz were small and non-significant for all TMG parameters. Conclusions: Clinically realistic PMT protocols at 35–45 Hz in non-fatigued muscle induce small but statistically detectable, duration-sensitive changes in TMG-derived contractile behavior over approximately 10 min. Within the constraints of the fixed, non-randomized design and the small effect sizes observed, these findings support viewing massage gun use as a recovery-oriented adjunct that subtly modulates contractile dynamics, rather than as a strong, standalone performance-enhancing stimulus. Full article

Petroleum hydrocarbons frequently contaminate arid oilfield soils, but remediation is challenging because these soils typically contain little organic matter, retain little moisture, and are exposed to high temperatures, that hinder natural attenuation. This study evaluated indigenous bioaugmentation of an aged crude oil-contaminated sandy soil from the Burgan oilfield in Kuwait, in contrast to exogenous commercial microbial products and to natural attenuation. In a 140-day bench-scale tray study, aged crude oil–contaminated soil from the Burgan oilfield (initial TPH 2.49–4.78%, dry wt.) was treated with an enriched indigenous consortium, a commercial consortium, or no inoculum under controlled moisture, nutrient, and aeration conditions. TPH was quantified as hexane-extractable material, and degradation kinetics were evaluated using a first-order model. A statistical comparison of replicate-derived decay constants (k) was conducted using one-way ANOVA and subsequent post hoc testing. Among the replicated treatments, the indigenous consortium showed the strongest performance. In the low-TPH indigenous group, TPH removal reached 63.8 ± 3.1% and fell below 1% by day 140; at higher starting TPH, removal remained substantial but slower. Commercial inoculation was less effective and more variable, while uninoculated controls showed minimal decline. The decay constant for the indigenous (0.0053–0.0075 day⁻¹) was much higher (p < 0.001) than those in commercial (0.0025 day⁻¹) and natural attenuation (0.0005 day⁻¹). Furthermore, the model fit was robust for indigenous treatments (R² = 0.89–0.91) but weaker for commercial and uninoculated controls. The study findings demonstrate that bioaugmentation utilizing well-adapted indigenous consortia offers a statistically validated and kinetically predictable strategy for TPH remediation in desert soils. Full article

Background/Objectives: To evaluate the associations between corneal topographic irregularity, higher-order aberrations (HOAs), and posterior segment structural and microvascular parameters in keratoconus using optical coherence tomography (OCT) and OCT angiography (OCTA). Methods: In this cross-sectional study, 81 eyes with keratoconus and 60 healthy control eyes underwent corneal topography and wavefront analysis, spectral-domain OCT with enhanced depth imaging, and OCTA. Retinal layer thicknesses, choroidal thickness and area metrics, choroidal vascularity index (CVI), and OCTA-derived vascular parameters were analyzed. Associations were assessed using Spearman correlation analysis with false discovery rate (FDR) correction. Results: Compared with controls, keratoconus eyes showed significantly increased corneal curvature, corneal irregularity indices, and HOAs (all p < 0.001). Structural OCT analysis demonstrated preserved inner retinal layers, whereas outer nuclear layer thickness was reduced (p < 0.001) and overall outer retinal layer thickness was increased (p = 0.005). Choroidal thickness and both total and luminal choroidal areas were significantly greater in keratoconus eyes (all p ≤ 0.011), while CVI did not differ between groups (p > 0.05). OCTA revealed reduced superficial capillary plexus vessel density at the whole image and perifoveal regions (all p < 0.001), whereas deep capillary plexus and foveal avascular zone metrics were largely preserved. Correlation analyses identified only weak and inconsistent associations between corneal parameters, HOAs, and posterior segment measurements, none of which remained statistically significant after FDR correction. Conclusions: Despite pronounced anterior segment deformation and optical degradation, posterior segment structural and microvascular alterations in keratoconus are limited and weakly related to corneal disease severity. These findings support a predominantly anterior segment centered pathophysiology of keratoconus and highlight the importance of stringent multiple-comparison control in multimodal imaging studies. Full article

This study quantifies how confinement changes the orientational phase space of proteins by comparing hanging-drop (HD) with Langmuir–Blodgett (LB) conditions within a unified probabilistic framework grounded in structural data from the Protein Data Bank (PDB). For each protein, principal moments of inertia are computed from atomic coordinates, trace-normalized, and used to define a geometry-based benchmark for the probability of occupying a predefined productive-orientation set. In parallel, a Hamiltonian-weighted probability is obtained within a classical statistical–mechanical treatment by reconstructing the orientational distribution over the polar–azimuthal domain under a fixed global confinement protocol. The analysis is carried out on a ten-protein panel spanning diverse sizes and anisotropies, and the HD→LB contrast is characterized through probability gains, distributional distances, and an energy-basin decomposition that distinguishes basin depth from basin measure. Under identical parameterization, LB globally produces higher productive-orientation probabilities than HD across all proteins, establishing a uniform direction of the confinement effect while preserving protein-dependent magnitudes. The inertia-based benchmark exhibits broader dispersion in LB/HD amplification, whereas the Hamiltonian construction yields a more regular cross-protein gain, consistent with LB acting as a global reweighting of orientational phase space rather than a protein-specific re-tuning. By integrating PDB-derived structural descriptors with a statistical–mechanical operator, the framework provides a transparent bridge between molecular geometry and confinement-driven ordering and offers a compact basis for comparing crystallization-relevant confinement protocols across structurally heterogeneous proteins. Full article

This paper develops a rigorous inferential framework for a class of Gaussian stochastic processes driven by white noise with constant drift, whose temporal evolution is governed by a Caputo fractional derivative of order ${\displaystyle \alpha \in (1/2,1)}$. The model belongs to the family of fractional Volterra processes, where memory is generated by the dynamics themselves rather than by correlated noise. We derive explicit analytical expressions for the mean, variance, and covariance structure of the solution, thereby characterizing in a precise manner how the fractional order ${\displaystyle \alpha }$ governs both variance growth and the strength of temporal dependence. In particular, the process exhibits correlated increments and a power-law variance scaling of order ${\displaystyle t^{2\alpha -1}}$, highlighting the dual role of ${\displaystyle \alpha }$ as a regularity and memory parameter. Building on this structural analysis, we address the statistical problem of estimating the parameter vector ${\displaystyle (\mu ,\sigma ,\alpha )}$ from discrete-time observations. Two complementary procedures are proposed for the estimation of the fractional order: a variance-growth method based on log–log regression of empirical variances, and a wavelet-based estimator exploiting multi-scale scaling properties of the process. For the drift and diffusion parameters ${\displaystyle (\mu ,\sigma )}$, we construct explicit Gaussian pseudo-maximum likelihood estimators derived from the Volterra covariance structure of the increment process. We establish unbiasedness, ${\displaystyle L^2}$-convergence, strong consistency, and asymptotic normality for all estimators. Furthermore, we derive Berry–Esseen type bounds that quantify the rate of convergence toward the Gaussian law, providing sharp distributional approximations in a genuinely fractional and non-Markovian setting. A Monte Carlo study is carried out, using high-resolution Volterra discretizations, large-scale simulation budgets, covariance-structured linear algebra, and multi-scale diagnostic tools. The numerical experiments confirm the theoretical convergence rates, demonstrate the finite-sample reliability of the estimators, and illustrate the sensitivity of the process dynamics to the fractional order ${\displaystyle \alpha }$: smaller values of ${\displaystyle \alpha }$ produce stronger memory effects and higher variability, while values closer to one lead to smoother and more stable trajectories. The proposed methodology unifies statistical inference for long-memory Gaussian processes with fractional differential stochastic dynamics, offering a coherent analytical and computational framework applicable in areas such as quantitative finance, anomalous diffusion in physics, hydrology, and engineering systems with hereditary effects. Full article

From an environmental perspective, the use of clam shells contributes positively to marine waste management and promotes more sustainable construction practices. This study aims to analyze the influence of clam shell-derived filler on the mechanical properties of cementitious mortars, evaluating its effect as a function of dosage and particle fineness, in order to determine its potential as a sustainable additive in construction applications. The shells were ground for 0.5, 1.0, and 1.5 h and incorporated at percentages ranging from 0.5% to 5.0% by mass of cement. Slump (reduced Abram’s cone) was performed in the fresh state for each specimen mixture, while flexural strength, and compressive strength tests were performed at 7, 14, and 28 days of curing. Microstructural characterization was also performed using scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDX) analysis. In addition, particle size distribution parameters were determined to quantify the effect of grinding time on particle refinement and its relationship with mechanical performance. A multifactor ANOVA was conducted to evaluate the statistical significance of grinding time and filler dosage on compressive strength. The results showed that the combination of 0.5 h of grinding and 1.0% filler provided the best mechanical performance for both flexural and compressive strength, with values of 7.27 MPa and 26.16 MPa, respectively. Dosages higher than 2.0% tended to decrease strength, which is associated with saturation of non-cementing particles. EDX analysis showed adequate calcium distribution without generating chemical segregation. The results showed that the combination of 0.5 h of grinding and 1.0% filler provided the best mechanical performance for both flexural and compressive strength, with values of 7.27 MPa and 26.16 MPa, respectively. Dosages higher than 2.0% tended to decrease strength, which is associated with saturation effects and increased specific surface area. The statistical analysis confirmed that both grinding time and filler dosage significantly influence compressive strength, highlighting the importance of optimizing particle size distribution and filler content to achieve improved mechanical performance. Full article

Background: Transient ischemic attack (TIA) and minor stroke often result in excellent functional recovery but are frequently followed by substantial psychological morbidity. It remains unclear whether mood disturbances or cognitive impairment are the primary contributors to reduced health-related quality of life (HRQoL) in this population. Methods: We conducted a prospective observational case–control study including 90 patients with acute TIA or minor stroke confirmed by diffusion-weighted imaging and 92 age-matched healthy controls. At 90 days, participants completed the Hamilton Depression Rating Scale, Hamilton Anxiety Rating Scale, Montreal Cognitive Assessment, and the EQ-5D-5L. Hierarchical multiple regression using standardized z-scores identified independent predictors of HRQoL. Bias-corrected bootstrapped mediation analyses (5000 iterations) assessed whether cognitive impairment mediated the relationship between mood symptoms and HRQoL. Results: Compared with controls, patients exhibited markedly higher rates of depressive symptoms (82.2% vs. 18.5%), anxiety symptoms (81.1% vs. 21.7%), and cognitive impairment (66.7% vs. 13.0%) (all p < 0.001). Psychopathological variables explained an additional 36.6% of HRQoL variance, whereas cognitive and neuroimaging variables contributed only 1.7% (ΔR² = 0.017; p = 0.523). In the fully adjusted regression model, HAM-A showed the numerically largest standardized coefficient (β = −0.055; p = 0.064), representing a trend toward significance, while HDRS-17 did not individually reach statistical significance (β = −0.043; p = 0.147); cognitive impairment had negligible independent effects (β = −0.001; p = 0.947). Both mood variables collectively accounted for the substantial majority of explained HRQoL variance, far exceeding the contribution of cognitive and neuroimaging predictors. Mediation analyses revealed no significant indirect effects, indicating that mood and cognitive complications are statistically consistent with a model in which mood and cognitive symptoms exert independent effects on HRQoL; temporal ordering cannot be established from these cross-sectional measures. Conclusions: Following TIA or minor stroke, depressive and anxiety symptoms are highly prevalent, persist despite good neurological recovery, and exert a disproportionately negative impact on HRQoL. Anxiety appears particularly influential in determining patient-reported outcomes. The statistical consistency of the mediation models with parallel rather than sequential mood–cognition pathways suggests that these represent independent neurobiological sequelae requiring separate clinical attention, underscoring the need for routine and concurrent assessment of both mood and cognitive function after TIA and minor stroke. Full article

Background: Physical activity requiring self-control may yield greater post-activity cognitive improvements. Implementing such tasks within The Daily Mile could enhance cognition further while adding variety to the initiative. This study examined whether the inclusion of self-control tasks within The Daily Mile influences subsequent cognition and enjoyment. Methods: Participants, numbering 99 (10.2 ± 1.1 y), completed three trials (Daily Mile Normal, Daily Mile Self-Control, and resting), using a within-subject, order-balanced, crossover design. The Daily Mile Self-Control involved students completing tasks requiring self-control for 30 s every 2 min within The Daily Mile. Cognitive tests (Stroop test, Sternberg Paradigm, Visual Search test) were administered prior to, immediately following and 45 min following The Daily Mile and resting trials. During the trials, distance covered (m), average heart rate (beats·min⁻¹) and physical activity enjoyment (PACES) were measured. Focus groups explored factors affecting enjoyment during The Daily Mile trials. The effects of physical activity vs. rest on cognitive function were examined first, followed by the effect of adding self-control tasks to The Daily Mile. Results: There were no statistically significant differences between the Daily Mile trials on distance covered or physical activity enjoyment. However, average heart rate was significantly higher in The Daily Mile Self-Control compared to The Daily Mile Normal. Compared to rest, The Daily Mile had a positive effect on inhibitory control. Working-memory accuracy maintained following activity, but perceptual accuracy was briefly impaired. Following the addition of self-control tasks, working memory response times improved. However, accuracy on inhibitory control and perception declined immediately after activity, compared to The Daily Mile Normal. Thematic analysis indicated varied perceptions among participants, with some valuing the simplicity and control of The Daily Mile Normal, and others favoring The Daily Mile Self-Control due to the variety and cognitive challenge. Conclusions: Incorporating self-control tasks into The Daily Mile produced mixed cognitive and qualitative responses, compared to The Daily Mile Normal. This suggests that tailoring physical activity to individual preferences may optimize engagement and cognitive outcomes. Full article

Objectives: The reduction in intrinsic capacity significantly impacts the functional abilities of older individuals, and is strongly linked to adverse health consequences. Safeguarding and enhancing an elderly person’s intrinsic capacity can lead to better life quality and improved social well-being. This research seeks to explore the relationships between social engagement, physical activity, and the likelihood of decline in intrinsic capacity among the elderly in China. Methods: Utilizing the CHARLS data from 2015, individuals with incomplete information were removed from our study. Our analysis included a total of 3502 samples. Social participation and physical activity were assessed through self-reported surveys. The evaluation of intrinsic capacity, based on WHO criteria, thoroughly examined participants in five areas: mobility, sensory functions, vitality, mental health and cognitive abilities. The links between social participation, physical activity and intrinsic capacity decline were revealed through logistic regression. Restricted cubic splines (RCS) were employed as a statistical model, exploring the relationships between dose and response. Interaction analysis was used to examine the interaction between social participation and physical activity. Analyses of subgroups facilitated the evaluation of variations based on factors including age, gender, duration of sleep, and chronic disease numbers. Results: In contrast to the low-level group, individuals with moderate to high degrees of social participation (OR = 0.80, p = 0.012; OR = 0.56, p < 0.001) and those with moderate to high levels physical activity (OR = 0.72, p = 0.019; OR = 0.74, p = 0.016) demonstrated a notably lower risk of decline in intrinsic capacities. A negative correlation was identified in a dose-response manner between social participation and the risk of IC decline. A U-shaped relationship was established between physical activity levels and the risk of intrinsic capacity decline. The fully adjusted interaction model showed that no significant interaction was observed between social participation and physical activity (p = 0.778). Furthermore, subgroup analyses showed that these associations remained generally consistent across older adults of different age groups, genders, sleep duration, and numbers of chronic diseases. Conclusions: In order to slow down the deterioration of intrinsic capacity in older adults in China, it may be beneficial to focus on sustaining elevated levels of social participation and engaging in moderate physical activity. Higher levels of social participation are associated with a lower risk of experiencing a decline in intrinsic capacity, whereas both insufficient and excessive physical activity are associated with an increased risk of intrinsic capacity decline. Full article

Diagnostic disclosure is a complex communication task that requires learners to integrate interpersonal attunement, structured information delivery, and condition-specific reasoning in real time. We conducted a randomized controlled trial comparing conventional diagnostic communication training with the same training supplemented by an AI-supported adaptive virtual patient simulation designed to provide additional deliberate practice and individualized, just-in-time feedback. Eighty undergraduate medical students were randomized 1:1 and completed standardized-patient encounters involving disclosure of a new diagnosis of type 2 diabetes mellitus before and after training. Performance was assessed by blinded physician raters using an adapted Kalamazoo rubric. Among students with complete pre–post data (conventional training, n = 25; AI-supported training, n = 26), both groups showed substantial improvement. Mean gains were numerically larger in the AI-supported group, with small-to-moderate standardized effects across selected communication domains; however, baseline-adjusted group-by-time interactions did not reach conventional thresholds for statistical significance, indicating that any added mean effects beyond conventional training remain uncertain. Exploratory person-level analyses suggested greater heterogeneity of improvement in the AI-supported condition, including a higher density of large gains in higher-order communication components. These findings should therefore be interpreted as exploratory rather than confirmatory. AI-supported adaptive simulation appears feasible as an adjunct to communication training, but adequately powered studies are needed to clarify effect magnitude, mechanisms, and generalizability across training contexts. Full article

Background: Artificial intelligence (AI) is increasingly incorporated into higher education. However, most empirical studies focus on technological adoption or learner satisfaction rather than on how pedagogical design, perceived learning impact, and student experience interact within AI-mediated learning environments. Understanding these relationships is essential to determine whether AI supports higher-order reasoning processes rather than merely increasing technological engagement. Objective: This preliminary study aimed to develop and evaluate a theory-driven AI-mediated pedagogical framework and examine relationships between pedagogical design, perceived learning impact, and student satisfaction in a university learning context. Methods: An observational educational evaluation was conducted during implementation of an AI-mediated instructional framework in an undergraduate physiotherapy course. The full academic cohort (n = 22) completed a 24-item questionnaire assessing seven pedagogical domains on a 5-point Likert scale. Descriptive statistics, Pearson correlations, exploratory regression modeling, and factor analysis were used to examine relationships among domains. Academic performance indicators were summarized descriptively. Results: Students reported high evaluations across all domains (means > 4.5/5). The strongest association with satisfaction was perceived learning impact (r = 0.79, p < 0.001). Moderate correlations were found for usability (r = 0.66), AI content quality (r = 0.61), pedagogical coherence (r = 0.58), critical thinking (r = 0.52), and ethical integration (r = 0.47). Academic pass rates exceeded 90%. Conclusions: Perceived learning impact emerged as the central mechanism linking AI-mediated instructional design to student satisfaction, suggesting that the educational value of AI depends on alignment with cognitively demanding learning processes. Full article

Gene transcription is inherently stochastic, and promoter-switching-induced transcriptional bursting generates substantial cell-to-cell variability in mRNA abundance. Such variability is commonly characterized by the mean and variance; however, these low-order statistics fail to capture the geometric features of mRNA copy number distributions and may obscure mechanistic differences in promoter dynamics. In this work, we analyze a two-state stochastic gene transcription model and derive explicit analytical expressions for higher-order moments of mRNA abundance. We show that skewness and kurtosis provide mechanistically informative signatures of transcriptional bursting, explicitly depending on promoter switching kinetics and burst size. Our results demonstrate that distinct promoter dynamics can produce identical mean expression levels and variances while exhibiting markedly different skewness and kurtosis. The explicit analytical expressions derived here reveal how higher-order moments encode mechanistically informative signatures of transcriptional bursting through distributional asymmetry and heavy-tailed behavior. These results demonstrate that higher-order moments encode mechanistic information beyond mean–variance statistics and provide a powerful framework for distinguishing between different promoter-switching mechanisms in stochastic gene transcription. Full article

Although plant restoration is essential for improving soil structure and stability, there are still few systematic assessments of its impacts across various restored vegetation species, especially in environmentally sensitive areas like the East Qinling Mountains. In order to provide a scientific foundation for optimizing restoration tactics and enhancing soil erosion control and ecosystem services in the area, this study attempts to assess the impacts of different recovered plant types on soil aggregate stability and to clarify the underlying mechanisms. The Pinus tabuliformis Carrière, Quercus variabilis Blume, Robinia pseudoacacia L., Pinus tabulaeformis-Quercus variabilis mixed forest, Platycladus orientalis (L.) Franco and abandoned grassland were the six vegetation types represented by the sixteen plots. Farmland was used as a control. Soil samples were taken from three depths (0–5 cm, 5–20 cm, and 20–40 cm) and evaluated for root biomass, soil organic matter (SOM), and water-stable aggregate dispersion. Mean weight diameter (MWD), fractal dimension (D), macroaggregate content of diameter > 0.25 mm (R_0.25), and percentage of aggregate disruption (PAD) were used to evaluate aggregate stability. One-way ANOVA, LSD multiple comparisons, and Spearman correlation analysis were among the statistical analyses. In comparison to grassland and farming, forested regions, particularly mixed forests, showed considerably higher proportions of macroaggregates (>0.25 mm) and superior aggregate stability (higher MWD and R_0.25, lower D and PAD). Increased litter and coarse root inputs, which encouraged big water-stable aggregates (WSAs) and reinforced their positive connection with SOM, were the driving forces behind this development. Robinia pseudoacacia L. and Platycladus orientalis (L.) Franco displayed the highest SOM concentration and root biomass (1201.45 and 679.66 g/m², respectively). At all depths, mixed forests showed the most stable soil structure. In contrast to agriculture, vegetation restoration dramatically changed the mechanical composition of the soil, increasing the differentiation of particle-size fractions across soil layers and decreasing the amount of surface clay. Soil aggregate stability is greatly enhanced by vegetation restoration, with mixed forests offering the greatest advantages because of their varied root systems and increased input of organic matter. These results emphasize how crucial it is to choose the right vegetation types for restoration efforts in order to improve soil structure, reduce erosion, and promote ecological sustainability in the East Qinling Mountains. Full article