University of Illinois Urbana-Champaign

Advances in functional magnetic resonance imaging (fMRI) have led to the ability to study the brain across many contexts. However, the large number of features generated by functional connectivity approaches may overfit the data. These problems can be overcome with either feature selection (FS) or dimensionality reduction (DR), which can be applied to less complex models. We utilize two open source datasets to compare the performance of DR/FS methods on cognitive task decoding using a suite of ML classifiers. While FS and DR methods have been used previously in decoding research, no systematic comparison of their performance has been undertaken. Here, we compare available methods using commonly utilized machine learning libraries to establish which methods provide the best predictive performance. We then conduct statistical tests to examine the relative contributions of DR and FS methods and classifiers on decoding accuracy. Neither DR or FS was found to be superior. However, differences were identified across datasets and tasks. In the majority of methods and datasets, a peak in predictive performance was found using a small percentage (005-.10%) of the total number of original features. Some methods perform better than the baseline method of prediction with all available features or selecting features randomly. Decoding performance with some datasets with some method exceeds that of deep learning approaches. These results suggest a "sweet spot" for the tradeoff between the retention of features and predictive accuracy.

Real-time simulation and the integration of digital twins necessitate the development of computationally efficient surrogate models for contemporary food drying systems, in which complex physical processes must interact dynamically with virtual representations. Traditional modeling approaches for food drying are subject to notable limitations: empirical models often lack physical fidelity, whereas high-fidelity computational fluid dynamics (CFD) simulations require the resolution of tightly coupled heat and mass transfer processes among vapor, liquid water, and air phases within unsaturated porous media, rendering them computationally prohibitive for real-time applications. This work proposes a region-aware graph neural network (GNN) surrogate model for food drying simulation by leveraging mesh-based and region-specific representations of the drying process. The proposed GNN architecture, inspired by MeshGraphNet with specialized masked message passing, handles the heterogeneous domains and sharp discontinuities characteristic of coupled heat and mass transfer in food drying. The GNN model was trained on a limited dataset comprising 28 CFD simulations of hot-air drying of a potato slice under varying ambient conditions, generated using a CFD framework validated against numerical results from the literature and laboratory-scale experimental data. Despite this small training set, the model delivers predictive performance for operating conditions beyond those seen during training. Single-step predictions achieved RMSE values of 0.17 °C for temperature ( T ), 0.26% for relative humidity ( RH ), and 0.0028 for liquid saturation ( S w ). Comparative analysis with standard message passing revealed a threefold reduction in interfacial absolute error (0.003 vs. 0.008) and elimination of nonphysical field propagation across domain boundaries. The model maintained performance on the training domain edges, with moisture ratio curves achieving R 2 > 0.97 for 10-step rollouts (R 2 = 0.989 interior domain; R 2 = 0.979 edge domain) while preserving the characteristic S-shaped drying curves in longer autonomous simulations (R 2 = 0.784 and 0.873 for full rollouts of the interior and edge-of-domain test cases, respectively). This framework is applicable to real-time process control, design optimization, and digital twin implementations in food manufacturing. • Integrated a region-aware GNN into a digital twin for potato drying. • Achieved high accuracy with minimal data in heat and mass transfer prediction. • Region-aware message passing ensured physical consistency across domains. • Model adapted to changes in temperature and geometry with low prediction error. • Fast predictions enable real-time control and food process optimization

The objective of this scoping review is to map the evidence on using smartwatches to objectively measure physical activity (PA), sedentary behavior (SB), and sleep in children, adolescents, adults, and older adults. This review followed the Joanna Briggs Institute (JBI) guidelines for scoping reviews, following a previously published protocol. The searches were conducted in January 2022 on Medline, Scopus, Web of Science, IEEE Xplore Digital Library, Scielo, LILACS, Health Technology Assessment Database, Cochrane clinical trials, and clinical trials. The screening was performed independently by two authors. A narrative synthesis was used. After the electronic search, 5,925 records were identified and 2,008 duplicates removed. Screening of 3,917 titles and abstracts resulted in 491 full-text articles assessed for eligibility. A total of 427 studies were excluded for not meeting inclusion criteria, with one study added after author contact. Thus, 64 of these studies were included. Sample sizes ranged widely from 4 to 6,454 participants, with most studies (80.9%) including 100 or fewer participants. Most studies were conducted in the United States (46.9%), followed by China (9.4%). A total of 18 different smartwatch brands were identified, with Apple® being the most investigated (46.9%), followed by Samsung® (15.6), Fitbit® (14.1%), Motorola® (10.9%), Polar® (10.9%), Garmin® (9.4%), Asus® (3.1%), and others (1.6% each. Smartwatches were primarily used to validate measurements of PA, SB, and/or sleep parameters, followed by their use in assessing these variables as exposures or outcomes, testing feasibility, supporting self-monitoring, and other purposes. The most frequently measured variables were steps, total sleep time, and SB. This review can help studies, interventions, and health professionals that aim to use this technology as a measurement instrument, as well as help end users who use it on a daily basis.

Growing interest in the career preferences of dark personalities has led to empirical investigation into the relationship between Dark Triad traits and vocational interests. However, little research has examined this relationship at the facet level. This study uses network analysis to investigate the relationships between seven facets of Dark Triad traits (psychopathic Boldness, Meanness, and Disinhibition; Machiavellian View and Tactics; and Narcissistic Admiration and Rivalry) and vocational interests. Undergraduate students ( N = 609) completed facet-level measures of the Dark Triad and, two weeks later, a vocational interest inventory assessing SETPOINT ( Health S cience, Creative E xpression, T echnology, P eople, O rganization, I nfluence, N ature, and T hings ) interest dimensions. Network analysis revealed differential relationships of Dark Triad facets to SETPOINT interests. In addition, network centrality (expected influence and strength) and bridge centrality indices (bridge expected influence) were computed. Influence (interest in leading and influencing others in business, politics, and legal domains) emerged as the most central variable, and the bridge node linking Dark Triad and vocational interest constructs. This implies that dark personality facets are most likely to co-occur with interest in leadership positions, especially in business and politics. Findings suggest network analysis as a promising approach for understanding the relationship between dark personality traits and vocational interests. • Network analysis examined Dark Triad facet–vocational interest relationships. • Interests were conceptualized using the SETPOINT model. • Patterns of partial correlations differed across different Dark Triad facets. • Influence interest dimension emerged as the most central node and the bridge node. • Findings provide nuanced insights into dark vocational preferences.

Hallucinated outputs from large language models (LLMs) pose risks in the medical domain, especially for lay audiences making health-related decisions. Existing automatic factual consistency evaluation methods, such as entailment- and question-answering (QA) -based, struggle with plain language summarization (PLS) due to elaborative explanation phenomenon, which introduces external content (e.g., definitions, background, examples) absent from the scientific abstract to enhance comprehension. To address this, we introduce PlainQAFact, an automatic factual consistency evaluation metric trained on a fine-grained, human-annotated dataset PlainFact, for evaluating factual consistency of both source-simplified and elaborately explained sentences. PlainQAFact first classifies sentence type, then applies a retrieval-augmented QA scoring method. Empirical results show that existing evaluation metrics fail to evaluate the factual consistency in PLS, especially for elaborative explanations, whereas PlainQAFact consistently outperforms them across all evaluation settings. We further analyze PlainQAFact's effectiveness across external knowledge sources, answer extraction strategies, answer overlap measures, and document granularity levels, refining its overall factual consistency assessment. Taken together, our work presents a sentence-aware, retrieval-augmented metric targeted at elaborative explanations in biomedical PLS tasks, providing the community with both a new benchmark and a practical evaluation tool to advance reliable and safe plain language communication in the medical domain. PlainQAFact and PlainFact are available at: https://github.com/zhiwenyou103/PlainQAFact.

The transition to college is associated with increased sleep difficulties. Everyday discrimination is an important social determinant of sleep difficulties, and this study examined associations between precollege exposures to everyday discrimination and sleep during the first year of college. The study also considers the mediating role of daily sleep quality. Using a combination of longitudinal surveys and daily diaries, the extent to which precollege exposures to everyday discrimination collected in the fall semester predicted sleep difficulties in the spring semester of students' first year in college was investigated in a sample of 628 diverse college students (mean age = 18.54 years). Sleep difficulties included self-reported sleep onset-related disturbance and daytime dysfunction. Precollege exposures to everyday discrimination were associated with higher levels of daytime dysfunction (but not with sleep disturbance) in the spring semester. Precollege exposures to everyday discrimination were also associated with lower mean levels of daily sleep quality across 14 days; in turn, lower daily sleep quality contributed to higher levels of daytime dysfunction in the spring. Although precollege exposures to everyday discrimination were associated with higher variability in daily sleep quality, variability in daily sleep quality was not associated with sleep difficulties in the spring. Precollege exposures to everyday discrimination were associated with sleep difficulties in the first year of college, and these sleep concerns are explained by daily-level sleep processes. Targeting daily sleep quality may be an important health lever to improve first-year college students' sleep problems and their transition to college. (PsycInfo Database Record (c) 2026 APA, all rights reserved).